Intelligent decision-making method and system for satellite intelligence support based on machine learning

Abstract

The application discloses an intelligent decision-making method and system for satellite intelligence support based on machine learning, and the method comprises the following steps: acquiring a satellite remote sensing image in satellite intelligence, inputting the satellite remote sensing image into a CNN neural network, and finding an identified target in the satellite remote sensing image; setting a shape feature extraction function, extracting the shape feature of the identified target, and setting a texture feature extraction function, extracting the texture feature of the identified target, wherein the shape feature of the identified target is extracted by performing a rotational symmetry operation on the identified target and calculating the height index of the identified target; combining the extracted shape feature and the texture feature, determining the morphology of the identified target, comparing the morphology with a preset object classification dictionary, thereby determining the object classification of the identified target, and making a decision according to the object classification of the identified target.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent decision-making based on satellite intelligence support, and more specifically, relates to an intelligent decision-making method and system based on satellite intelligence support based on machine learning. Background Art

[0002] The information obtained by satellites can be used to assist in various decision-making processes and can be applied in many fields, including military, agriculture, meteorology, environmental monitoring, etc. The following are some intelligent decision-making methods that use satellite intelligence to improve decision-making efficiency:

[0003] Image recognition and analysis: Target recognition and analysis using satellite imagery can provide critical information in military, agricultural, and environmental fields. Technologies such as deep learning can automatically identify and classify surface features such as buildings, vegetation, and water bodies, supporting decision-making.

[0004] Geographic Information System (GIS) analysis: Combining satellite imagery with other geographic data to create GIS for spatial analysis. This can help decision makers better understand geospatial relationships for planning and forecasting.

[0005] However, the accuracy of target recognition in existing technologies is not high, and the recognition results may have obvious deviations, thereby affecting the accuracy of decision-making. Summary of the Invention

[0006] To solve the above technical problems, the present invention proposes an intelligent decision-making method supported by satellite intelligence based on machine learning, comprising:

[0007] Obtaining satellite remote sensing images from satellite intelligence, inputting the satellite remote sensing images into a CNN neural network, and finding identified targets in the satellite remote sensing images;

[0008] Setting a shape feature extraction function to extract the shape features of the identified target, and setting a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target;

[0009] The extracted shape features are combined with the texture features to determine the morphology of the identified target, and compared with a preset object classification dictionary to determine the object classification of the identified target, and a decision is made based on the object classification of the identified target.

[0010] Furthermore, the shape feature extraction function includes:

[0011] ,

[0012] ,

[0013] in, The coordinates of the identified target in the satellite remote sensing image The shape recognition function at is used to identify the contour of the identified target, is the altitude index of the identified target, used to determine the altitude of the identified target. The coordinates of the identified target in the satellite remote sensing image The shape evolution function at is used to identify the shape of the identified target.

[0014] Furthermore, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0015] ,

[0016] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0017] Furthermore, extracting texture features of the identified target includes:

[0018] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0019] Furthermore, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0020] ,

[0021] The first The transformed pixel value With threshold Compare and generate Binary results ,if Greater than or equal to threshold , then the binary result is set to 1, otherwise it is set to 0, specifically:

[0022] ,

[0023] According to Binary results , set the texture feature extraction function , extract the texture of the identified target, specifically:

[0024] ,

[0025] in, For coordinates The pixel value at For coordinates The pixel value at .

[0026] The present invention also proposes an intelligent decision-making system supported by satellite intelligence based on machine learning, comprising:

[0027] A data acquisition module is used to acquire satellite remote sensing images from satellite intelligence, input the satellite remote sensing images into a CNN neural network, and find identified targets in the satellite remote sensing images;

[0028] a feature extraction module, configured to set a shape feature extraction function to extract the shape features of the identified target, and set a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target;

[0029] The decision module is used to combine the extracted shape features with the texture features to determine the morphology of the identified target, and compare it with the preset object classification dictionary to determine the object classification of the identified target, and make a decision based on the object classification of the identified target.

[0030] Furthermore, the shape feature extraction function includes:

[0031] ,

[0032] ,

[0033] in, The coordinates of the identified target in the satellite remote sensing image The shape recognition function at is used to identify the contour of the identified target, is the altitude index of the identified target, used to determine the altitude of the identified target. The coordinates of the identified target in the satellite remote sensing image The shape evolution function at is used to identify the shape of the identified target.

[0034] Furthermore, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0035] ,

[0036] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0037] Furthermore, extracting texture features of the identified target includes:

[0038] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0039] Furthermore, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0040] ,

[0041] The first The transformed pixel value With threshold Compare and generate Binary results ,if Greater than or equal to threshold , then the binary result is set to 1, otherwise it is set to 0, specifically:

[0042] ,

[0043] According to Binary results , set the texture feature extraction function , extract the texture of the identified target, specifically:

[0044] ,

[0045] in, For coordinates The pixel value at For coordinates The pixel value at .

[0046] Compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0047] The present invention obtains satellite remote sensing images from satellite intelligence, inputs the satellite remote sensing images into a CNN neural network, and finds the identified targets in the satellite remote sensing images; sets a shape feature extraction function to extract the shape features of the identified targets, and sets a texture feature extraction function to extract the texture features of the identified targets, wherein extracting the shape features of the identified targets includes: performing a rotational symmetry operation on the identified targets and calculating the height index of the identified targets; combining the extracted shape features with the texture features to determine the morphology of the identified targets, and comparing them with a preset object classification dictionary to determine the object classification of the identified targets, and making decisions based on the object classification of the identified targets. Through the above technical solutions, the present invention can identify identified targets through satellite intelligence data, perform object classification, and make corresponding decisions based on the object classification results. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0049] Figure 2 This is a system structure diagram of Example 2 of the present invention. DETAILED DESCRIPTION

[0050] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0051] The method provided by the present invention can be implemented in the following terminal environment, wherein the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0052] A processor can include one or more processing cores. It connects various components within the terminal using various interfaces and circuits. It executes instructions, programs, code sets, or instruction sets stored in storage media, and accesses data stored in storage media to perform various terminal functions and process data.

[0053] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets, or instructions.

[0054] The display screen is used to display the interactive interface of each application.

[0055] All subscripts in the formulas of the present invention are only used to distinguish parameters and have no actual meaning.

[0056] In addition, those skilled in the art will appreciate that the structure of the terminal described above does not limit the terminal. The terminal may include more or fewer components, or a combination of certain components, or a different arrangement of components. For example, the terminal may also include a radio frequency circuit, an input unit, a sensor, an audio circuit, a power supply, and other components, which will not be described in detail here.

[0057] Example 1

[0058] like Figure 1 As shown, an embodiment of the present invention provides an intelligent decision-making method supported by satellite intelligence based on machine learning, including:

[0059] Step 101: obtaining a satellite remote sensing image from satellite intelligence, inputting the satellite remote sensing image into a CNN neural network, and finding an identified target in the satellite remote sensing image;

[0060] Step 102: Setting a shape feature extraction function to extract the shape features of the identified target, and setting a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target;

[0061] Specifically, the shape feature extraction function includes:

[0062] ,

[0063] ,

[0064] in, The coordinates of the identified target in the satellite remote sensing image The shape recognition function at is used to identify the contour of the identified target, is the altitude index of the identified target, used to determine the altitude of the identified target. The coordinates of the identified target in the satellite remote sensing image The shape evolution function at is used to identify the shape of the identified target.

[0065] Specifically, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0066] ,

[0067] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0068] Specifically, extracting texture features of the identified target includes:

[0069] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0070] Specifically, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0071] ,

[0072] The first The transformed pixel value With threshold Compare and generate Binary results ,if Greater than or equal to threshold , then the binary result is set to 1, otherwise it is set to 0, specifically:

[0073] ,

[0074] According to Binary results , set the texture feature extraction function , extract the texture of the identified target, specifically:

[0075] ,

[0076] wherein, is a pixel value at coordinate , is a pixel value at coordinate .

[0077] Step 103, combining the extracted shape feature and the texture feature, determining the morphology of the identified target, and comparing with a preset object classification dictionary, so as to determine the object classification of the identified target, and making a decision according to the object classification of the identified target.

[0078] For example, the object classification can be battlefield vehicles such as airplanes, artillery, tanks, etc., when battlefield simulation training is carried out, if the identified target is an airplane, the ground personnel can be decided to hide in the air-raid shelter or bunker, so as to ensure the personal safety of the ground personnel, if the object classification of the identified target is artillery, the decision can be made to carry out point destruction.

[0079] Embodiment 2

[0080] As shown in Figure 2 , the embodiment of the present application also proposes an intelligent decision system of satellite intelligence support based on machine learning, comprising:

[0081] The data acquisition module is used for acquiring satellite remote sensing images in satellite intelligence, inputting the satellite remote sensing images into the CNN neural network, and finding out the identified target in the satellite remote sensing images.

[0082] The feature extraction module is used for setting a shape feature extraction function, extracting the shape feature of the identified target, and setting a texture feature extraction function, extracting the texture feature of the identified target, wherein the shape feature extraction of the identified target comprises: performing a rotational symmetry operation on the identified target, and calculating the height index of the identified target.

[0083] Specifically, the shape feature extraction function comprises:

[0084] ,

[0085] ,

[0086] wherein, is a shape identification function of the identified target on the satellite remote sensing image at coordinate , used for identifying the contour of the identified target, is the height index of the identified target, used for judging the height condition of the identified target, is a texture identification function of the identified target on the satellite remote sensing image at coordinate The shape evolution function at is used to identify the shape of the identified target.

[0087] Specifically, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0088] ,

[0089] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0090] Specifically, extracting texture features of the identified target includes:

[0091] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0092] Specifically, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0093] ,

[0094] The first The transformed pixel value With threshold Compare and generate Binary results ,if Greater than or equal to threshold , then the binary result is set to 1, otherwise it is set to 0, specifically:

[0095] ,

[0096] According to Binary results , set the texture feature extraction function , extract the texture of the identified target, specifically:

[0097] ,

[0098] in, For coordinates The pixel value at For coordinates The pixel value at .

[0099] The decision module is used to combine the extracted shape features with the texture features to determine the morphology of the identified target, and compare it with the preset object classification dictionary to determine the object classification of the identified target, and make a decision based on the object classification of the identified target.

[0100] Example 3

[0101] An embodiment of the present invention also proposes a storage medium storing a plurality of instructions, wherein the instructions are used to implement the intelligent decision-making method supported by satellite intelligence based on machine learning.

[0102] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0103] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: Step 101, obtaining a satellite remote sensing image from satellite intelligence, inputting the satellite remote sensing image into a CNN neural network, and finding an identified target in the satellite remote sensing image;

[0104] Step 102: Setting a shape feature extraction function to extract the shape features of the identified target, and setting a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target;

[0105] Specifically, the shape feature extraction function includes:

[0106] ,

[0107] ,

[0108] in, The coordinates of the identified target in the satellite remote sensing image The shape recognition function at is used to identify the contour of the identified target, is the altitude index of the identified target, used to determine the altitude of the identified target. The coordinates of the identified target in the satellite remote sensing image The shape evolution function at is used to identify the shape of the identified target.

[0109] Specifically, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0110] ,

[0111] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0112] Specifically, extracting texture features of the identified target includes:

[0113] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0114] Specifically, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0115] ,

[0116] The first The transformed pixel value With threshold Compare and generate Binary results ,if Greater than or equal to threshold , then the binary result is set to 1, otherwise it is set to 0, specifically:

[0117] ,

[0118] According to Binary results , set the texture feature extraction function , extract the texture of the identified target, specifically:

[0119] ,

[0120] wherein, is a pixel value at coordinate is a pixel value at coordinate is a pixel value at coordinate is a pixel value at coordinate

[0121] Step 103, combine the extracted shape features and texture features to determine the morphology of the identified target, and compare with the preset object classification dictionary, so as to determine the object classification of the identified target, and make a decision according to the object classification of the identified target.

[0122] Embodiment 4

[0123] The embodiment of the application also provides an electronic device, including a processor and a storage medium connected with the processor, the storage medium stores a plurality of instructions, the instructions can be loaded and executed by the processor, so that the processor can execute the intelligent decision-making method of satellite intelligence support based on machine learning.

[0124] Specifically, the electronic device of the embodiment can be a computer terminal, which can include one or more processors and a storage medium.

[0125] The storage medium can be used to store software programs and modules, such as the intelligent decision-making method of satellite intelligence support based on machine learning in the embodiment of the application, and the corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, implements the intelligent decision-making method of satellite intelligence support based on machine learning. The storage medium can include a high-speed random storage medium, and can also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memories, or other non-volatile solid-state storage media. In some examples, the storage medium can further include storage media remotely arranged relative to the processor, which can be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0126] The processor can call the information and application programs stored in the storage medium through the transmission system to execute the following steps: step 101, acquiring a satellite remote sensing image in satellite intelligence, inputting the satellite remote sensing image into a CNN neural network, and finding an identified target in the satellite remote sensing image;

[0127] Step 102: Setting a shape feature extraction function to extract the shape features of the identified target, and setting a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target;

[0128] Specifically, the shape feature extraction function includes:

[0129] ,

[0130] ,

[0131] in, The coordinates of the identified target in the satellite remote sensing image The shape recognition function at is used to identify the contour of the identified target, is the altitude index of the identified target, used to determine the altitude of the identified target. The coordinates of the identified target in the satellite remote sensing image The shape evolution function at is used to identify the shape of the identified target.

[0132] Specifically, a rotational symmetry operation is performed on the identified target, and the height index of the identified target is calculated. include:

[0133] ,

[0134] in, is the first adjustment factor for controlling the amplitude of the function, is the polar angle, is the degree of rotational symmetry, is the second adjustment factor for controlling the amplitude of the function, is the polar diameter, is the adjustment factor used to control the width of the surface.

[0135] Specifically, extracting texture features of the identified target includes:

[0136] Define a circular neighborhood with a radius of R, with coordinates The pixel point at is taken as the center pixel point, and the pixel coordinate set within the circular neighborhood is obtained. ,in , is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed.

[0137] Specifically, transforming the pixel values ​​of all pixels within the circular neighborhood includes:

[0138] ,

[0139] The transformed pixel value in the circular neighborhood is compared with a threshold value, and a first binary result is generated. If the transformed pixel value is greater than or equal to the threshold value, the binary result is set to 1, otherwise, it is set to 0, specifically:

[0140] ,

[0141] According to the first binary result, a texture feature extraction function is set, and texture extraction is performed on the identified target, specifically:

[0142] ,

[0143] Wherein, is a pixel value at coordinate, and is a pixel value at coordinate.

[0144] Step 103: Combine the extracted shape feature and texture feature to determine the morphology of the identified target, and compare it with a preset object classification dictionary to determine the object classification of the identified target, and make a decision according to the object classification of the identified target.

[0145] The above-mentioned embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0146] In the above-mentioned embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0147] ​​​​​​​​​​​​​​In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, and can be electrical or other forms.

[0148] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0149] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0150] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, and other media that can store program code.

[0151] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. An intelligent decision-making method supported by satellite intelligence based on machine learning, characterized in that: include: Obtaining satellite remote sensing images from satellite intelligence, inputting the satellite remote sensing images into a CNN neural network, and finding identified targets in the satellite remote sensing images; Setting a shape feature extraction function to extract the shape features of the identified target, and setting a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target; The shape feature extraction function includes: Wherein, GC(x, y) is the shape recognition function of the identified target at the coordinate (x, y) on the satellite remote sensing image, which is used to identify the outline of the identified target; z is the height index of the identified target, which is used to determine the height of the identified target; MF(x, y) is the shape evolution function of the identified target at the coordinate (x, y) on the satellite remote sensing image, which is used to identify the shape of the identified target; Performing a rotational symmetry operation on the identified target and calculating a height index z of the identified target includes: Where A is the first adjustment factor for controlling the amplitude of the function, θ is the polar angle, n is the number of rotational symmetry, B is the second adjustment factor for controlling the amplitude of the function, r is the polar radius, and σ is the adjustment factor for controlling the width of the surface; Extracting texture features of the identified target includes: Define a circular neighborhood with a radius of R, take the pixel point at the coordinate (x, y) as the center pixel point, and obtain the pixel point coordinate set (x, y) in the circular neighborhood. k ,y k ), where k = 0, 1, ... P-1, P is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed; Transforming the pixel values ​​of all pixels within the circular neighborhood includes: x′ k =tanh(I(x k ,y k )-I(x,y)) The kth transformed pixel value x′ in the circular neighborhood k Compare with the threshold T to generate the kth binary result s k , if x′ k If it is greater than or equal to the threshold T, then the binary result s k is set to 1, otherwise it is set to 0, specifically: According to the kth binary result s k , set the texture feature extraction function LBP P，R (x, y), perform texture extraction on the identified target, specifically: Among them, I(x k ,y k ) is the coordinate (x k ,y k ), I(x, y) is the pixel value at coordinate (x, y); The extracted shape features are combined with the texture features to determine the morphology of the identified target, and compared with a preset object classification dictionary to determine the object classification of the identified target, and a decision is made based on the object classification of the identified target.

2. An intelligent decision-making system supported by satellite intelligence based on machine learning, characterized in that: include: A data acquisition module is used to acquire satellite remote sensing images from satellite intelligence, input the satellite remote sensing images into a CNN neural network, and find identified targets in the satellite remote sensing images; a feature extraction module, configured to set a shape feature extraction function to extract the shape features of the identified target, and set a texture feature extraction function to extract the texture features of the identified target, wherein extracting the shape features of the identified target includes: performing a rotational symmetry operation on the identified target and calculating a height index of the identified target; The shape feature extraction function includes: Wherein, GC(x, y) is the shape recognition function of the identified target at the coordinate (x, y) on the satellite remote sensing image, which is used to identify the outline of the identified target; z is the height index of the identified target, which is used to determine the height of the identified target; MF(x, y) is the shape evolution function of the identified target at the coordinate (x, y) on the satellite remote sensing image, which is used to identify the shape of the identified target; Performing a rotational symmetry operation on the identified target and calculating a height index z of the identified target includes: Where A is the first adjustment factor for controlling the amplitude of the function, θ is the polar angle, n is the number of rotational symmetry, B is the second adjustment factor for controlling the amplitude of the function, r is the polar radius, and σ is the adjustment factor for controlling the width of the surface; Extracting texture features of the identified target includes: Define a circular neighborhood with a radius of R, take the pixel point at the coordinate (x, y) as the center pixel point, and obtain the pixel point coordinate set (x, y) in the circular neighborhood. k ,y k ), where k = 0, 1, ... P-1, P is the number of pixels in the circular neighborhood, and the pixel values ​​of all pixels in the circular neighborhood are transformed; Transforming the pixel values ​​of all pixels within the circular neighborhood includes: x′ k =tanh(I(x k ,y k )-I(x,y)) The kth transformed pixel value x′ in the circular neighborhood k Compare with the threshold T to generate the kth binary result s k , if x′ k If it is greater than or equal to the threshold T, then the binary result s k is set to 1, otherwise it is set to 0, specifically: According to the kth binary result s k , set the texture feature extraction function LBP P，R (x, y), perform texture extraction on the identified target, specifically: Among them, I(x k ,y k ) is the coordinate (x k ,y k ), I(x, y) is the pixel value at coordinate (x, y); The decision module is used to combine the extracted shape features with the texture features to determine the morphology of the identified target, and compare it with the preset object classification dictionary to determine the object classification of the identified target, and make a decision based on the object classification of the identified target.

Images