A method for identifying corn-soybean strip intercropping areas

Abstract

This application discloses a method for identifying corn-soybean strip intercropping areas. The method includes: acquiring a set of satellite remote sensing images corresponding to the corn-soybean strip intercropping areas within a preset corn-soybean growth cycle; preprocessing the satellite remote sensing images to obtain a preprocessed first satellite remote sensing image set; extracting optical and texture features from the first satellite remote sensing image set, calculating the total value of the separability index (SI) corresponding to the optical features, and extracting principal component components based on the texture features; determining a second satellite remote sensing image set based on the total SI value and principal component components, and segmenting it to obtain a third satellite remote sensing image set after superpixel segmentation; and analyzing the second and third satellite remote sensing image sets to obtain the distribution of corn-soybean strip intercropping areas. This application solves the technical problem that existing corn-soybean intercropping area identification schemes cannot meet practical needs.

Description

Technical Field

[0001] This application relates to the field of agricultural remote sensing technology, and in particular to a method for identifying corn-soybean strip intercropping areas. Background Technology

[0002] Corn and soybeans are important crops that integrate food, feed, and processing. The corn-soybean strip intercropping model has significant ecological benefits and is recognized worldwide as the best intercropping model. This model effectively reduces energy consumption, carbon and nitrogen emissions, mitigates continuous cropping obstacles, and improves soil fertility through technologies such as no-till straw mulching, root nodule nitrogen fixation, and strip rotation.

[0003] Identifying soybean-maize intercropping using satellite remote sensing can fully leverage the wide coverage and rich spectral information of satellite imagery data. Combined with relevant agricultural remote sensing identification algorithms, it can identify the spatial distribution of soybean-maize intercropping, thereby increasing crop yields, farmers' incomes, and providing strong support for ensuring national food production and security.

[0004] Because the corn-soybean intercropping model has only recently begun to be widely adopted, there is currently no mature remote sensing method in the industry that can be used for production to identify corn-soybean intercropping areas. Furthermore, due to the complexity and unique characteristics of the corn-soybean intercropping model, existing extraction methods designed for other crop growing areas cannot be readily applied to the identification of corn-soybean intercropping areas. Summary of the Invention

[0005] The technical problem addressed by this application is that existing methods for identifying corn-soybean intercropping areas cannot meet practical needs. This application provides a method for identifying corn-soybean strip intercropping areas. The method provided in this application involves extracting optical and texture features from a set of satellite remote sensing images, calculating the total value of the separability index (SI) corresponding to the optical features, and extracting principal component components based on a preset principal component analysis method and the texture features. A second set of satellite remote sensing images is determined based on the total value of the SI and the principal component components. The second set of satellite remote sensing images is then segmented to obtain a third set of satellite remote sensing images after superpixel segmentation. The distribution of corn-soybean strip intercropping areas is obtained by analyzing the second and third sets of satellite remote sensing images. In other words, this method identifies corn-soybean strip intercropping areas based on the fusion of optical and texture features from satellite remote sensing images and superpixel segmentation, effectively ensuring highly automated identification and accurate extraction of corn-soybean intercropping areas.

[0006] In a first aspect, embodiments of this application provide a method for identifying corn-soybean strip intercropping areas. The method includes: acquiring a set of satellite remote sensing images corresponding to corn-soybean strip intercropping areas within a preset corn-soybean growth cycle; preprocessing the satellite remote sensing image set to obtain a preprocessed first satellite remote sensing image set; extracting optical features and texture features from the first satellite remote sensing image set, calculating the total value of the separability index (SI) corresponding to the optical features, and extracting principal component components based on a preset principal component analysis method and the texture features; determining a second satellite remote sensing image set based on the total value of the SI and the principal component components; segmenting the second satellite remote sensing image set to obtain a third satellite remote sensing image set after superpixel segmentation; and analyzing the second satellite remote sensing image set and the third satellite remote sensing image set to obtain the distribution of corn-soybean strip intercropping areas.

[0007] Optionally, preprocessing the satellite remote sensing image set to obtain a preprocessed first satellite remote sensing image set includes: performing geometric correction and radiometric correction on the satellite remote sensing images in the satellite remote sensing image set to obtain a corrected satellite remote sensing image set; and performing median synthesis on the satellite remote sensing images in the corrected satellite remote sensing image set to obtain the first satellite remote sensing image set.

[0008] Optionally, calculating the total value of the separability index SI corresponding to the optical feature includes: extracting samples of each land cover in the first satellite remote sensing image set to obtain a sample set; calculating the value of the separability index SI between any two land covers in the sample set for the optical feature; and superimposing the values ​​of the separability index SI corresponding to the optical feature to obtain the total value of the separability index SI.

[0009] Optionally, the optical characteristics include: green, red, blue and near-infrared bands, and spectral indices such as Enhanced Vegetation Index (EVI), Soil Modified Vegetation Index (MSAVI), Soil Optimized Vegetation Index (OSAVI), Transformed Chlorophyll Absorption Reflectance Index (TCARI), and Wide Range Dynamic Vegetation Index (WDRVI).

[0010] Optionally, the texture features include: calculating a gray-level co-occurrence matrix based on a gray-level image generated from green, red, blue, and near-infrared bands, and obtaining contrast, angular second moment, correlation, mean, variance, homogeneity, dissimilarity, and entropy co-occurrence based on the gray-level co-occurrence matrix.

[0011] Optionally, the separability index SI of the optical feature between any two land features in the sample set is calculated, including:

[0012] The separability index SI for each optical feature between any two land cover types in the sample set is calculated using the following formula:

[0013]

[0014] Among them, SI ij (m, n) represents the separability index SI of the m-th optical feature between land cover i and land cover j at time n; m represents the number of extracted optical features; n represents the time when the satellite remote sensing image was acquired; i and j represent any two land cover types in the sample set, respectively. and σ represents the average spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time; i and σ j The standard deviation of the spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time is represented.

[0015] Optionally, extracting ground feature samples from the first satellite remote sensing image set to obtain a sample set includes: expanding the ground feature samples in the first satellite remote sensing image set according to the recorded geographic coordinate information of the soybean-maize intercropping strip area to obtain an expanded satellite remote sensing image set; and extracting each ground feature sample from the expanded satellite remote sensing image set to obtain the sample set.

[0016] Optionally, determining a second satellite remote sensing image set based on the total value of the SI and the principal component components includes: determining the sensitivity information of each optical feature according to the total value of the SI corresponding to each optical feature; linearly combining the remote sensing images in the first satellite remote sensing image set according to the sensitivity information and the principal component components to obtain a fourth satellite remote sensing image set; and selecting a second satellite remote sensing image set that meets preset conditions from the fourth satellite remote sensing image set using a preset random forest classifier.

[0017] Optionally, segmenting the second satellite remote sensing image set to obtain a third satellite remote sensing image set after superpixel segmentation includes: performing clustering calculations on the satellite remote sensing images in the second satellite remote sensing image set using a preset clustering algorithm to obtain clustered satellite remote sensing images; and performing superpixel segmentation on the clustered satellite remote sensing images based on preset segmentation parameters to obtain the third satellite remote sensing image set.

[0018] Optionally, analyzing the second and third satellite remote sensing image sets to obtain the distribution of maize-soybean strip intercropping areas includes: constructing a stacked classification model, using the second and third satellite remote sensing image sets as input to the stacked classification model, and analyzing the second and third satellite remote sensing image sets using the stacked classification model to obtain the distribution of maize-soybean strip intercropping areas.

[0019] Secondly, this application provides a computer device, the computer device comprising:

[0020] Memory, used to store at least one instruction executed by a processor;

[0021] A processor is configured to execute instructions stored in memory to perform the method described in the first aspect.

[0022] Thirdly, this application provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the method described in the first aspect.

[0023] Compared with the prior art, the embodiments of this application have at least the following beneficial effects:

[0024] The solution provided in this application extracts optical and texture features from a set of satellite remote sensing images, calculates the total value of the separability index (SI) corresponding to the optical features, and extracts principal component components based on a preset principal component analysis method and the texture features. A second set of satellite remote sensing images is determined based on the total value of the SI and the principal component components. The second set of satellite remote sensing images is then segmented to obtain a third set of satellite remote sensing images after superpixel segmentation. Analysis of the second and third sets of satellite remote sensing images reveals the distribution of corn-soybean strip intercropping areas. In other words, the corn-soybean strip intercropping areas are identified based on the fusion of optical and texture features from satellite remote sensing images and superpixel segmentation, effectively ensuring highly automated identification and accurate extraction of corn-soybean intercropping areas. Attached Figure Description

[0025] Figure 1 A flowchart illustrating a method for identifying corn-soybean strip intercropping areas provided in this application embodiment;

[0026] Figure 2 This is a schematic diagram illustrating a strip-planting method for intercropping corn and soybeans, as provided in an embodiment of this application.

[0027] Figure 3 This is a schematic diagram of a corn-soybean strip intercropping area provided in an embodiment of this application;

[0028] Figure 4 This application provides an example of an extraction result image of a corn-soybean strip intercropping area based on a superpixel segmentation and classification algorithm model.

[0029] Figure 5 A flowchart illustrating another method for identifying corn-soybean strip intercropping areas provided in this application embodiment;

[0030] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0031] The embodiments described in this application are only a part of the embodiments, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0032] To better understand the above technical solutions, the technical solutions of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of this application and the specific features in the embodiments are detailed descriptions of the technical solutions of this application, rather than limitations on the technical solutions of this application. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other.

[0033] The following description, in conjunction with the accompanying drawings, provides a more detailed explanation of a method for identifying corn-soybean strip intercropping areas provided in this application. The specific implementation of this method may include the following steps (method flow as follows): Figure 1 As shown):

[0034] Step 101: Obtain a set of satellite remote sensing images corresponding to the corn-soybean strip intercropping area within the preset corn-soybean growth cycle, and preprocess the set of satellite remote sensing images to obtain a preprocessed first set of satellite remote sensing images.

[0035] Since soybeans and corn cannot be grown year-round, and each has a fixed growth cycle in different regions, satellite remote sensing images within the growth cycle of soybeans and corn are needed to identify soybean-corn strip intercropping areas. With the rapid development of satellite technology, the types of satellites are also increasing. For example, the satellite remote sensing images provided in this application embodiment are remote sensing images acquired by the Gaofen (GF) series satellites; that is, the satellite remote sensing images provided in this application embodiment are GF remote sensing images.

[0036] Soybean and corn can be planted in various ways, including but not limited to planting corn alone, planting soybeans alone, and intercropping soybeans and corn, such as intercropping corn and soybeans in strip planting (e.g., corn and soybeans in strip planting). Figure 2 (As shown). For example, a corn-soybean strip intercropping area refers to an area where corn-soybean strip intercropping exists and / or where corn or soybeans are grown separately, for example, such as... Figure 3This illustration shows a schematic diagram of a corn-soybean strip intercropping area provided in an embodiment of this application. Furthermore, in order to identify the corn-soybean strip intercropping area, the acquired satellite remote sensing images need to be selected not only from images within the corn-soybean growth cycle but also from images showing the presence of the corn-soybean strip intercropping area.

[0037] Furthermore, during the generation of remote sensing images, atmospheric transmission effects and remote sensing imaging characteristics can lead to inconsistencies between the geometric position, shape, and orientation of the satellite remote sensing images and the features intended to be represented by the system, resulting in low-quality images. To improve the quality of satellite remote sensing images, preprocessing is necessary to obtain high-quality images. Preprocessing includes geometric correction and radiometric correction. For example, geometric and radiometric corrections are performed on the satellite remote sensing images in the image set to obtain a corrected image set; median composite analysis is then performed on the corrected image set to obtain the first satellite remote sensing image set.

[0038] Step 102: Extract optical features and texture features from the first satellite remote sensing image set, calculate the total value of the separability index SI corresponding to the optical features, and extract principal component components according to the preset principal component analysis method and the texture features.

[0039] Satellite remote sensing imagery includes various features, such as geometric features, optical features, and textural features. For example, the optical features extracted from the first satellite remote sensing image set include: green, red, blue, and near-infrared bands; spectral indices include Enhanced Vegetation Index (EVI), Modified Soil Adjusted Vegetation Index (MSAVI), Optimized Soil-Adjusted Vegetation Index (OSAVI), Transformed Chlorophyll Absorption Ratio Index (TCARI), and Wide Dynamic Range (WD RVI).

[0040] For example, spectral indices include enhanced vegetation indices, soil-improved vegetation indices, soil-optimized vegetation indices, convertible chlorophyll absorption reflectance indices, and wide-range dynamic vegetation indices, as detailed below:

[0041]

[0042]

[0043]

[0044]

[0045]

[0046] Among them, NIR represents the near-infrared band; Red represents the red wave band; Blue represents the blue wave band; and Green represents the green wave band.

[0047] As another example, the texture features extracted from the first satellite remote sensing image set include: calculating the gray-level co-occurrence matrix based on grayscale images generated from green, red, blue, and near-infrared bands; and obtaining eight texture features based on the gray-level co-occurrence matrix, namely contrast, angular second moment, correlation, mean, variance, homogeneity, dissimilarity, and entropy co-occurrence; the grayscale image calculation formula is as follows:

[0048] Gray=(0.3×NIR)+(0.59×Red)×(0.11×Green)

[0049] Furthermore, as an example, calculating the total value of the separability index SI corresponding to the optical feature includes: extracting each land cover sample from the first satellite remote sensing image set to obtain a sample set; calculating the value of the separability index SI between any two land covers in the sample set for the optical feature; and superimposing the values ​​of each separability index SI corresponding to the optical feature to obtain the total value of the separability index SI.

[0050] As another example, calculating the separability index SI of the optical feature between any two land features in the sample set includes: calculating the separability index SI of each optical feature between any two land features in the sample set using the following formula:

[0051]

[0052] Among them, SI ij (m, n) represents the separability index SI of the m-th optical feature between land cover i and land cover j at time n; m represents the number of extracted optical features; n represents the time when the satellite remote sensing image was acquired; i and j represent any two land cover types in the sample set, respectively. and σ represents the average spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time; i and σ jThe standard deviation of the spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time is represented.

[0053] The satellite remote sensing image may contain more than one ground feature sample; for example, the ground feature samples may include corn, soybeans, water bodies, and corn-soybean intercropping. As an example, the ground feature samples in the first satellite remote sensing image set are expanded based on the recorded geographic coordinate information of the corn-soybean intercropping strip area to obtain an expanded satellite remote sensing image set; each ground feature sample is extracted from the expanded satellite remote sensing image set to obtain the sample set. For example, the recorded geographic coordinate information of the corn-soybean intercropping strip area is obtained using GPS. Another example is that each ground feature sample is segmented, with 70% used for training and the remaining 30% used for validation.

[0054] Furthermore, the extracted texture and optical features are ranked by feature importance. For optical features, the separability index (SI) is calculated across different land cover samples, and the scores are accumulated and ranked to determine sensitivity information. For texture features, multiple principal component components are obtained using principal component analysis. Then, based on the sensitivity information and the principal component components from low to high, a recursive feature elimination strategy based on random forest is applied to determine the optimal feature sets for both optical and texture features.

[0055] As an example, the sensitivity information of each optical feature is determined based on the total value of the SI corresponding to each optical feature; the remote sensing images in the first satellite remote sensing image set are linearly combined based on the sensitivity information and the principal component components to obtain a fourth satellite remote sensing image set; a preset random forest classifier is used to select a second satellite remote sensing image set from the fourth satellite remote sensing image set that meets preset conditions. For example, the preset condition is that the accuracy meets a preset threshold.

[0056] Step 103: Determine the second satellite remote sensing image set based on the total value of SI and the principal component components, and segment the second satellite remote sensing image set to obtain the third satellite remote sensing image set after superpixel segmentation.

[0057] As an example, segmenting the second satellite remote sensing image set to obtain a third satellite remote sensing image set after superpixel segmentation includes: performing clustering calculations on the satellite remote sensing images in the second satellite remote sensing image set using a preset clustering algorithm to obtain clustered satellite remote sensing images; and performing superpixel segmentation on the clustered satellite remote sensing images based on preset segmentation parameters to obtain the third satellite remote sensing image set. For example, the Simple Non-Iterative Clustering (SNIC) algorithm is used to segment optical feature images to obtain superpixel segmentation results; the preset segmentation parameters include compactness, connectivity, neighborhood size, and seeds; for example, compactness = 0, connectivity = 8, neighborhood size = 256, and seeds = 50. Specifically, such as... Figure 4 The image shown is an extraction result of a corn-soybean strip intercropping area based on a superpixel segmentation and classification algorithm model provided in an embodiment of this application.

[0058] Step 104: Analyze the second satellite remote sensing image set and the third satellite remote sensing image set to obtain the distribution of the corn-soybean strip intercropping area.

[0059] As an example, the distribution of corn-soybean strip intercropping areas is obtained by analyzing the second and third satellite remote sensing image sets, including: constructing a stacked classification model, using the second and third satellite remote sensing image sets as input to the stacked classification model, and analyzing the second and third satellite remote sensing image sets through the stacked classification model to obtain the distribution of corn-soybean strip intercropping areas.

[0060] For example, based on the selected second satellite remote sensing image set and the superpixel segmentation results (the aforementioned third satellite remote sensing image set), a stacking classification model is constructed by combining three machine learning algorithms: CatBoost, LightGBM, and Random Forest. Specifically, the stacking classification model consists of two levels: the first level uses a combination of minimum distance, classification regression tree, support vector machine, and random forest to form a base classifier, thereby reducing overfitting and outputting preliminary prediction results; the second level uses only random forest as a meta-classifier to further learn the feature information from the first level and output the final result. Figure 5 This document presents a flowchart illustrating another method for identifying corn-soybean strip intercropping areas provided in an embodiment of this application.

[0061] Furthermore, the distribution results of soybean-maize strip intercropping areas were obtained based on the aforementioned Stacking classification model. In addition, after obtaining the distribution results, accuracy evaluation was performed. For example, accuracy evaluation indicators included producer accuracy (PA), user accuracy (UA), overall accuracy (OA), and the Kappa coefficient.

[0062] For example, the accuracy PA, user accuracy UA, overall accuracy OA, and Kappa coefficients are shown below:

[0063]

[0064]

[0065]

[0066]

[0067] Among them, C xx C represents the number of correctly classified samples, corresponding to the x-th column of the x-th row in the confusion matrix. +x and C x+ These are the sums of all samples in the x-th column and x-th row of the confusion matrix, respectively. N is the total number of validation samples, and p is the number of columns in the confusion matrix.

[0068] The solution provided in this application extracts optical and texture features from a set of satellite remote sensing images, calculates the total value of the separability index (SI) corresponding to the optical features, and extracts principal component components based on a preset principal component analysis method and the texture features. A second set of satellite remote sensing images is determined based on the total value of the SI and the principal component components. The second set of satellite remote sensing images is then segmented to obtain a third set of satellite remote sensing images after superpixel segmentation. Analysis of the second and third sets of satellite remote sensing images reveals the distribution of corn-soybean strip intercropping areas. In other words, the corn-soybean strip intercropping areas are identified based on the fusion of optical and texture features from satellite remote sensing images and superpixel segmentation, effectively ensuring highly automated identification and accurate extraction of corn-soybean intercropping areas.

[0069] See Figure 6 This application provides a computer device, the computer device comprising:

[0070] Memory 601 is used to store at least one instruction executed by a processor;

[0071] Processor 602 is used to execute instructions stored in memory. Figure 1 The method described.

[0072] This application provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform... Figure 1 The method described.

[0073] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0074] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0075] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0076] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0077] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for identifying a corn-soybean belt composite planting area, characterized in that, include: Obtain a set of satellite remote sensing images corresponding to the strip intercropping area of ​​corn and soybean within a preset corn-soybean growth cycle, and preprocess the set of satellite remote sensing images to obtain a preprocessed first set of satellite remote sensing images. The satellite remote sensing image set is preprocessed to obtain a preprocessed first satellite remote sensing image set, including: Geometric and radiometric corrections are performed on the satellite remote sensing images in the aforementioned satellite remote sensing image set to obtain the corrected satellite remote sensing image set; The first satellite remote sensing image set is obtained by performing median synthesis on the satellite remote sensing images in the corrected satellite remote sensing image set. Optical and texture features are extracted from the first satellite remote sensing image set, the total value of the separability index SI corresponding to the optical features is calculated, and principal component components are extracted according to the preset principal component analysis method and the texture features. The second satellite remote sensing image set is determined based on the total value of SI and the principal component components, and the second satellite remote sensing image set is segmented to obtain the third satellite remote sensing image set after superpixel segmentation. The total value of the separability index SI corresponding to the optical features is calculated, including: A sample set is obtained by extracting samples of various land features from the first satellite remote sensing image set, and the separability index SI of the optical feature between any two land features in the sample set is calculated. The total value of the separability index SI is obtained by superimposing the values ​​of each separability index SI corresponding to the optical feature; Calculating the separability index SI of the optical feature between any two land features in the sample set includes: The separability index SI for each optical feature between any two land cover types in the sample set is calculated using the following formula: in, The value of SI represents the separability index between ground feature i and ground feature j at time n for the m-th optical feature; m represents the number of extracted optical features; n represents the time when the satellite remote sensing image was acquired; i and j represent any two ground features in the sample set, respectively. and This represents the average spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time. and The standard deviation of the spectral characteristics of any two land features at the corresponding satellite remote sensing image acquisition time; A sample set is obtained by extracting ground feature samples from the first satellite remote sensing image set, including: The land cover samples in the first satellite remote sensing image set are expanded based on the recorded geographic coordinate information of the soybean-corn intercropping strip area to obtain the expanded satellite remote sensing image set. The sample set is obtained by extracting samples of various ground features from the expanded satellite remote sensing image set; The second satellite remote sensing image set is determined based on the total value of the SI and the principal component components, including: The sensitivity information of each optical feature is determined based on the total value of the SI corresponding to each optical feature; Based on the sensitivity information and the principal component, the remote sensing images in the first satellite remote sensing image set are linearly combined to obtain the fourth satellite remote sensing image set. A preset random forest classifier is used to select a second set of satellite remote sensing images that meet preset conditions from the fourth set of satellite remote sensing images; The distribution of corn-soybean strip intercropping areas was obtained by analyzing the second and third satellite remote sensing image sets.

2. The method as described in claim 1, characterized in that, The optical characteristics include: green, red, blue and near-infrared bands, and the spectral indices are Enhanced Vegetation Index (EVI), Soil Modified Vegetation Index (MSAVI), Soil Optimized Vegetation Index (OSAVI), Transformed Chlorophyll Absorption Reflectance Index (TCARI), and Wide Range Dynamic Vegetation Index (WDRVI).

3. The method as described in claim 1, characterized in that, The texture features include: calculating a gray-level co-occurrence matrix based on gray-level images generated from green, red, blue, and near-infrared bands, and obtaining contrast, angular second moment, correlation, mean, variance, homogeneity, dissimilarity, and entropy co-occurrence based on the gray-level co-occurrence matrix.

4. The method as described in claim 1, characterized in that, The second satellite remote sensing image set is segmented to obtain a third satellite remote sensing image set after superpixel segmentation, including: A preset clustering algorithm is used to perform clustering calculations on the satellite remote sensing images in the second satellite remote sensing image set to obtain clustered satellite remote sensing images; The third satellite remote sensing image set is obtained by superpixel segmentation of the clustered satellite remote sensing images based on preset segmentation parameters.

5. The method as described in claim 1, characterized in that, Analysis of the second and third satellite remote sensing image sets revealed the distribution of maize-soybean strip intercropping areas, including: A stacked classification model is constructed, with the second satellite remote sensing image set and the third satellite remote sensing image set as inputs to the stacked classification model. The distribution of maize-soybean strip intercropping areas is obtained by analyzing the second satellite remote sensing image set and the third satellite remote sensing image set through the stacked classification model.

Images