A method and system for target feature extraction and classification recognition in SAR images
By performing quality correction and multi-scale feature extraction on SAR images and combining the target decision model of environmental factors, the problem of insufficient building recognition accuracy in SAR images is solved and high-precision building recognition is achieved.
Patent Information
- Application Number
- CN202411970337.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-09-30
- Estimated Expiration
- 2044-12-30
AI Technical Summary
Building recognition in SAR images is affected by noise or interference, resulting in insufficient recognition accuracy.
By setting the image quality correction function and combining the noise information to correct the original SAR image, a new SAR image is generated. The target feature vector is extracted using the multi-scale feature mapping function. The target decision model is set in combination with environmental factors to determine whether the target to be identified is a building.
The recognition accuracy of buildings in SAR images is improved, and buildings can be accurately identified.
Smart Images
Figure CN120088637B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of image recognition, and more specifically, relates to a method and system for extracting and classifying target features in SAR images. Background Art
[0002] SAR (Synthetic Aperture Radar) imaging is a technology that uses radar waves to image the ground. SAR imaging works based on synthetic aperture radar technology, which involves transmitting microwave signals via radar systems on satellites or aircraft and receiving the reflected echoes. Because radar waves can penetrate clouds and form high-resolution images of the ground, SAR imaging technology is currently widely used in areas such as building monitoring.
[0003] Although SAR imaging technology has been widely used in the field of building monitoring, due to the influence of some noise or other interference sources, the recognition of buildings in SAR images will be biased. Therefore, there is an urgent need for a technical solution to improve the accuracy of building recognition. Summary of the Invention
[0004] To solve the above technical problems, the present invention proposes a method for target feature extraction and classification recognition in SAR images for identifying buildings, comprising:
[0005] Acquire an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity;
[0006] Dividing the new SAR image into multiple scales according to resolution, setting a multi-scale feature mapping function, and extracting a multi-scale feature vector of the target to be identified at the position (x, y);
[0007] A target decision model is set to calculate the target recognition value of the target to be identified based on the multi-scale feature vector and combined with environmental factors, and to determine whether the target to be identified is a building, wherein the environmental factors include: temperature, humidity, and air pressure.
[0008] Furthermore, the target decision model includes:
[0009]
[0010] Among them, C physical (I, x, y) is the target recognition value of the target to be identified at position (x, y) on the new SAR image I, M is the number of environmental factors, η mis the weight of the mth environmental factor, α2 is the adjustment factor of the multi-scale feature mapping function, F multi-scale is the multi-scale feature mapping function, γ m is the first adjustment factor of the mth environmental factor, λ m is the second adjustment factor of the mth environmental factor, Λ m (x, y) is the mth environmental factor at position (x, y) on the original SAR image I′;
[0011] When the target recognition value C of the target to be recognized at position (x, y) on the new SAR image I is physical (I, x, y) is greater than the preset building threshold τ building , confirm that the target to be identified is a building.
[0012] Furthermore, the multi-scale feature mapping function F mmlti-scale include:
[0013]
[0014] Where S is the number of scales, α s is the first adjustment factor of the sth scale, γ s is the second adjustment factor of the sth scale, I(x, y) is the image quality correction function at position (x, y) on the original SAR image I′, and a new SAR image I is generated, μ s is the third adjustment factor of the sth scale, γ s is the fourth adjustment factor of the s-th scale, is the height gradient at position (x, y) on the new SAR image I, η s is the fifth adjustment factor of the sth scale, δ s is the sixth adjustment factor of the sth scale, T(x, y) is the temperature at position (x, y) on the original SAR image I, T0 is the standard temperature, H(x, y) is the humidity at position (x, y) on the original SAR image I, and H0 is the standard humidity.
[0015] Furthermore, the image quality correction function I(x, y) at the position (x, y) on the original SAR image I′ includes:
[0016]
[0017] Among them, I terrain (x, y) is the terrain influence function at position (x, y) on the original SAR image I′, which is used to describe the influence of terrain on target feature extraction, η a′ is the adjustment factor of the noise source, (x0, y0) is the center coordinate of the noise source, σ noiseis the standard deviation of the noise intensity.
[0018] Furthermore, the terrain influence function I at position (x, y) on the original SAR image I′ is terrain (x, y) includes:
[0019]
[0020] Among them, I enhanced (x, y) is the reflection characteristic and propagation loss function at the position (x, y) on the original SAR image I′, which is used to describe the reflection characteristic and propagation loss of the radar wave when the original SAR image I′ is taken. t is the first adjustment factor of the terrain influence function, β t is the second adjustment factor of the terrain influence function, γ m′ is the third adjustment factor of the terrain influence function, M′(x, y) is the terrain feature vector at position (x, y) on the original SAR image I′, is the height gradient at position (x, y) on the original SAR image I′.
[0021] Furthermore, the reflection characteristics and propagation loss function I at position (x, y) on the original SAR image I′ are enhanced (x, y) includes:
[0022] I enhanced (x, y)
[0023] =[I′·exp(-α r ·d(x,y))]·[1+β e R(x, y)]
[0024] ·exp(-β a″ A(x, y)) (1+γ f ·exp(-α v ·v′ p ))
[0025] Among them, I′ is the original SAR image, α r is the distance adjustment factor, d(x, y) is the distance between the target at position (x, y) on the original SAR image I′ and the synthetic aperture radar, β e is the cross-section adjustment factor, R(x, y) is the cross-section of the synthetic aperture radar at position (x, y) on the original SAR image I′, β a″ is the adjustment factor of air pressure, A(x, y) is the air pressure at position (x, y) on the original SAR image I′, γ f is the first adjustment factor of the radar wave frequency, α vis the second adjustment factor of the frequency of the radar wave, v′ is the frequency of the radar wave, and p is the third adjustment factor of the frequency of the radar wave.
[0026] The present invention also proposes a target feature extraction and classification recognition system in SAR images for identifying buildings, comprising:
[0027] An image correction module is configured to obtain an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity;
[0028] A feature extraction module is used to divide the new SAR image into multiple scales according to the resolution, and set a multi-scale feature mapping function to extract the multi-scale feature vector of the target to be identified at the position (x, y);
[0029] The judgment module is used to set a target decision model, calculate the target recognition value of the target to be identified based on the multi-scale feature vector and combined with environmental factors, and determine whether the target to be identified is a building, wherein the environmental factors include: temperature, humidity, and air pressure.
[0030] Furthermore, the target decision model includes:
[0031]
[0032] Among them, C physical (I, x, y) is the target recognition value of the target to be identified at position (x, y) on the new SAR image I, M is the number of environmental factors, η m is the weight of the mth environmental factor, α2 is the adjustment factor of the multi-scale feature mapping function, F multi-scale is the multi-scale feature mapping function, γ m is the first adjustment factor of the mth environmental factor, λ m is the second adjustment factor of the mth environmental factor, Λ m (x, y) is the mth environmental factor at position (x, y) on the original SAR image I′;
[0033] When the target recognition value C of the target to be recognized at position (x, y) on the new SAR image I is physical (I, x, y) is greater than the preset building threshold τ building , confirm that the target to be identified is a building.
[0034] Furthermore, the multi-scale feature mapping function F multi-scale include:
[0035]
[0036] Where S is the number of scales, α s is the first adjustment factor of the sth scale, γ s is the second adjustment factor of the sth scale, I(x, y) is the image quality correction function at position (x, y) on the original SAR image I′, and a new SAR image I is generated, μ s is the third adjustment factor of the sth scale, γ s is the fourth adjustment factor of the s-th scale, is the height gradient at position (x, y) on the new SAR image I, η s is the fifth adjustment factor of the sth scale, δ s is the sixth adjustment factor of the sth scale, T(x, y) is the temperature at position (x, y) on the original SAR image I, T0 is the standard temperature, H(x, y) is the humidity at position (x, y) on the original SAR image I, and H0 is the standard humidity.
[0037] Furthermore, the image quality correction function I(x, y) at the position (x, y) on the original SAR image I′ includes:
[0038]
[0039] Among them, I terrain (x, y) is the terrain influence function at position (x, y) on the original SAR image I′, which is used to describe the influence of terrain on target feature extraction, η a′ is the adjustment factor of the noise source, (x0, y0) is the center coordinate of the noise source, σ noise is the standard deviation of the noise intensity.
[0040] In general, the above technical solutions conceived by the present invention have the following beneficial effects compared with the prior art:
[0041] The present invention uses an image quality correction function to correct the quality of SAR images, generates new SAR images, sets a multi-scale feature mapping function, extracts the multi-scale feature vector of the target to be identified, and finally sets a target decision model to calculate the target recognition value of the target to be identified and judge whether the target to be identified is a building. It can accurately identify buildings in SAR images. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;
[0043] Figure 2 is a system structure diagram of embodiment 2 of the present invention; DETAILED DESCRIPTION
[0044] 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.
[0045] 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.
[0046] 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.
[0047] 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.
[0048] The display is used to show the user interface of each application.
[0049] 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.
[0050] Example 1
[0051] like Figure 1 As shown, an embodiment of the present invention proposes a method for extracting and classifying target features in SAR images for identifying buildings, including:
[0052] Step 101: Acquire an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity;
[0053] Specifically, the image quality correction function I(x, y) at position (x, y) on the original SAR image I′ includes:
[0054]
[0055] Among them, Iterrain (x, y) is the terrain influence function at position (x, y) on the original SAR image I′, which is used to describe the influence of terrain on target feature extraction, η a′ is the adjustment factor of the noise source, (x0, y0) is the center coordinate of the noise source, σ noise is the standard deviation of the noise intensity.
[0056] Specifically, the terrain influence function I at position (x, y) on the original SAR image I′ is terrain (x, y) includes:
[0057]
[0058] Among them, I enhanced (x, y) is the reflection characteristic and propagation loss function at the position (x, y) on the original SAR image I′, which is used to describe the reflection characteristic and propagation loss of the radar wave when the original SAR image I′ is taken. t is the first adjustment factor of the terrain influence function, β t is the second adjustment factor of the terrain influence function, γ m′ is the third adjustment factor of the terrain influence function, M′(x, y) is the terrain feature vector at position (x, y) on the original SAR image I′, is the height gradient at position (x, y) on the original SAR image I′.
[0059] Specifically, the reflection characteristics and propagation loss function I at position (x, y) on the original SAR image I′ enhanced (x, y) includes:
[0060] I enhanced (x, y)
[0061] =[I′·exp(-α r ·d(x,y))]·[1+β e R(x, y)]
[0062] ·exp(-β a″ A(x, y)) (1+γ f ·exp(-α v ·v′ p ))
[0063] Among them, I′ is the original SAR image, α r is the distance adjustment factor, d(x, y) is the distance between the target at position (x, y) on the original SAR image I′ and the synthetic aperture radar, β eis the cross-section adjustment factor, R(x, y) is the cross-section of the synthetic aperture radar at position (x, y) on the original SAR image I′, β a″ is the adjustment factor of air pressure, A(x, y) is the air pressure at position (x, y) on the original SAR image I′, γ f is the first adjustment factor of the radar wave frequency, α v is the second adjustment factor of the frequency of the radar wave, v′ is the frequency of the radar wave, and p is the third adjustment factor of the frequency of the radar wave.
[0064] Step 102: Divide the new SAR image into multiple scales according to resolution, set a multi-scale feature mapping function, and extract a multi-scale feature vector of the target to be identified at the position (x, y);
[0065] Specifically, the multi-scale feature mapping function F multi-scale include:
[0066]
[0067] Where S is the number of scales, α s is the first adjustment factor of the sth scale, γ s is the second adjustment factor of the sth scale, I(x, y) is the image quality correction function at position (x, y) on the original SAR image I′, and a new SAR image I is generated, μ s is the third adjustment factor of the sth scale, γ s is the fourth adjustment factor of the s-th scale, is the height gradient at position (x, y) on the new SAR image I, η s is the fifth adjustment factor of the sth scale, δ s is the sixth adjustment factor of the sth scale, T(x, y) is the temperature at position (x, y) on the original SAR image I, T0 is the standard temperature, H(x, y) is the humidity at position (x, y) on the original SAR image I, and H0 is the standard humidity.
[0068] Step 103: Set a target decision model, calculate the target recognition value of the target to be identified based on the multi-scale feature vector and combined with environmental factors, and determine whether the target to be identified is a building, wherein the environmental factors include: temperature, humidity, and air pressure.
[0069] Specifically, the target decision model includes:
[0070]
[0071] Among them, C physical(I, x, y) is the target recognition value of the target to be identified at position (x, y) on the new SAR image I, M is the number of environmental factors, η m is the weight of the mth environmental factor, α2 is the adjustment factor of the multi-scale feature mapping function, F multi-scale is the multi-scale feature mapping function, γ m is the first adjustment factor of the mth environmental factor, λ m is the second adjustment factor of the mth environmental factor, Λ m (x, y) is the mth environmental factor at position (x, y) on the original SAR image I′;
[0072] When the target recognition value C of the target to be recognized at position (x, y) on the new SAR image I is physical (I, x, y) is greater than the preset building threshold τ building , confirm that the target to be identified is a building.
[0073] Example 2
[0074] like Figure 2 As shown, an embodiment of the present invention further provides a target feature extraction and classification recognition system in SAR images for identifying buildings, comprising:
[0075] An image correction module is configured to obtain an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity;
[0076] Specifically, the image quality correction function I(x, y) at position (x, y) on the original SAR image I′ includes:
[0077]
[0078] Among them, I terrainn (x, y) is the terrain influence function at position (x, y) on the original SAR image I′, which is used to describe the influence of terrain on target feature extraction, η a′ is the adjustment factor of the noise source, (x0, y0) is the center coordinate of the noise source, σ noise is the standard deviation of the noise intensity.
[0079] Specifically, the terrain influence function I at position (x, y) on the original SAR image I′ is terrain (x, y) includes:
[0080]
[0081] Among them, Ienhanced (x, y) is the reflection characteristic and propagation loss function at the position (x, y) on the original SAR image I′, which is used to describe the reflection characteristic and propagation loss of the radar wave when the original SAR image I′ is taken. t is the first adjustment factor of the terrain influence function, β t is the second adjustment factor of the terrain influence function, γ m ′ is the third adjustment factor of the terrain influence function, M′(x, y) is the terrain feature vector at position (x, y) on the original SAR image I′, is the height gradient at position (x, y) on the original SAR image I′.
[0082] Specifically, the reflection characteristics and propagation loss function I at position (x, y) on the original SAR image I′ enhanced (x, y) includes:
[0083] I enhanced (x, y)
[0084] =[I′·exp(-α r ·d(x,y))]·[1+β e R(x, y)]
[0085] ·exp(-β a″ A(x, y)) (1+γ f ·exp(-α v ·v′ p ))
[0086] Among them, I′ is the original SAR image, α r is the distance adjustment factor, d(x, y) is the distance between the target at position (x, y) on the original SAR image I′ and the synthetic aperture radar, β e is the cross-section adjustment factor, R(x, y) is the cross-section of the synthetic aperture radar at position (x, y) on the original SAR image I′, β a″ is the adjustment factor of air pressure, A(x, y) is the air pressure at position (x, y) on the original SAR image I′, γ f is the first adjustment factor of the radar wave frequency, α v is the second adjustment factor of the frequency of the radar wave, v′ is the frequency of the radar wave, and p is the third adjustment factor of the frequency of the radar wave.
[0087] A feature extraction module is used to divide the new SAR image into multiple scales according to the resolution, and set a multi-scale feature mapping function to extract the multi-scale feature vector of the target to be identified at the position (x, y);
[0088] Specifically, the multi-scale feature mapping function Fmulti-scale include:
[0089]
[0090] Where S is the number of scales, α s is the first adjustment factor of the sth scale, γ s is the second adjustment factor of the sth scale, I(x, y) is the image quality correction function at position (x, y) on the original SAR image I′, and a new SAR image I is generated, μ s is the third adjustment factor of the sth scale, γ s is the fourth adjustment factor of the s-th scale, is the height gradient at position (x, y) on the new SAR image I, η s is the fifth adjustment factor of the sth scale, δ s is the sixth adjustment factor of the sth scale, T(x, y) is the temperature at position (x, y) on the original SAR image I, T0 is the standard temperature, H(x, y) is the humidity at position (x, y) on the original SAR image I, and H0 is the standard humidity.
[0091] The judgment module is used to set a target decision model, calculate the target recognition value of the target to be identified based on the multi-scale feature vector and combined with environmental factors, and determine whether the target to be identified is a building, wherein the environmental factors include: temperature, humidity, and air pressure.
[0092] Specifically, the target decision model includes:
[0093]
[0094] Among them, C physical (I, x, y) is the target recognition value of the target to be identified at position (x, y) on the new SAR image I, M is the number of environmental factors, η m is the weight of the mth environmental factor, α2 is the adjustment factor of the multi-scale feature mapping function, F multi-scale is the multi-scale feature mapping function, γ m is the first adjustment factor of the mth environmental factor, λ m is the second adjustment factor of the mth environmental factor, Λ m (x, y) is the mth environmental factor at position (x, y) on the original SAR image I′;
[0095] When the target recognition value C of the target to be recognized at position (x, y) on the new SAR image I is physical (I, x, y) is greater than the preset building threshold τ building , confirm that the target to be identified is a building.
[0096] Example 3
[0097] An embodiment of the present invention further provides a storage medium storing a plurality of instructions, wherein the instructions are used to implement the method for extracting and classifying target features in SAR images.
[0098] 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.
[0099] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the steps of Embodiment 1.
[0100] Example 4
[0101] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to perform the method for extracting and classifying target features in SAR images.
[0102] Specifically, the electronic device of this embodiment may be a computer terminal, which may include: one or more processors, and a storage medium.
[0103] The storage medium can be used to store software programs and modules, such as the method for extracting and classifying target features in SAR images and corresponding program instructions / modules, as described in an embodiment of the present invention. The processor executes the software programs and modules stored in the storage medium to perform various functional applications and data processing, thereby implementing the method for extracting and classifying target features in SAR images. The storage medium can include high-speed random access memory (RAM) and non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some embodiments, the storage medium can further include storage media located remotely from the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0104] The processor can call the information and application programs stored in the storage medium through the transmission system to execute the steps of Example 1.
[0105] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.
[0106] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0107] 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 merely a logical function division. In actual implementation, there may be other division methods, such as 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.
[0108] 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.
[0109] 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.
[0110] 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 perform 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, etc. Various media that can store program codes.
[0111] 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. A method for extracting and classifying target features in SAR images for identifying buildings, characterized in that: include: Acquire an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity; The new SAR image is divided into multiple scales according to the resolution, and a multi-scale feature mapping function is set to extract the position The multi-scale feature vector of the target to be identified; Setting a target decision model to calculate a target recognition value of the target to be identified based on the multi-scale feature vector and in combination with environmental factors, and determining whether the target to be identified is a building, wherein the environmental factors include temperature, humidity, and air pressure; The target decision model includes: ， ， in, For the new SAR image Upper position The target identification value of the target to be identified, is the number of environmental factors, For the The weight of environmental factors, is the adjustment factor of the multi-scale feature mapping function, is the multi-scale feature mapping function, For the The first adjustment factor for environmental factors is For the The second adjustment factor for environmental factors is In the original SAR image Upper position The first environmental factors; When the new SAR image Upper position The target recognition value of the target to be identified Greater than the preset building threshold When , it is confirmed that the target to be identified is a building; Multi-scale feature mapping function include: ， in, is the number of scales, For the The first adjustment factor for the scale, For the The second adjustment factor for each scale, In the original SAR image Upper position The image quality correction function at , and generate a new SAR image , For the The third adjustment factor of the scale, For the The fourth adjustment factor of the scale, For the new SAR image Upper position The height gradient at For the The fifth adjustment factor of the scale, For the The sixth adjustment factor of the scale, In the original SAR image Upper position The temperature at is the standard temperature, In the original SAR image Upper position The humidity of the The standard humidity.
2. The method for target feature extraction and classification in SAR images according to claim 1, wherein: In the original SAR image Upper position Image quality correction function at include: ， in, In the original SAR image Upper position The terrain influence function at is used to describe the influence of terrain on target feature extraction. is the adjustment factor of the noise source, is the center coordinate of the noise source, is the standard deviation of the noise intensity.
3. The method for target feature extraction and classification in SAR images according to claim 2, wherein: In the original SAR image Upper position Terrain influence function at include: ， in, In the original SAR image Upper position The reflection characteristics and propagation loss function at are used to describe the original SAR image. The reflection characteristics and propagation loss of radar waves, is the first adjustment factor of the terrain influence function, is the second adjustment factor of the terrain influence function, is the third adjustment factor of the terrain influence function, In the original SAR image Upper position The terrain feature vector at In the original SAR image Upper position The height gradient at .
4. The method for target feature extraction and classification in SAR images according to claim 3, wherein: In the original SAR image Upper position Reflection characteristics and propagation loss function at include: ， in, is the original SAR image, is the distance adjustment factor, In the original SAR image Upper position The distance between the target and the synthetic aperture radar, is the cross-section adjustment factor, In the original SAR image Upper position The cross section of the synthetic aperture radar at is the adjustment factor for air pressure, In the original SAR image Upper position The air pressure at is the first adjustment factor of the radar wave frequency, is the second adjustment factor for the frequency of the radar wave, is the frequency of the radar wave, is the third adjustment factor of the radar wave frequency.
5. A target feature extraction and classification recognition system in SAR images for identifying buildings, characterized in that: include: An image correction module is configured to obtain an original SAR image, set an image quality correction function, perform image quality correction on the original SAR image in combination with noise information, and generate a new SAR image, wherein the noise information includes: the center position of the noise source and the standard deviation of the noise intensity; Feature extraction module, used to divide the new SAR image into multiple scales according to resolution, and set a multi-scale feature mapping function to extract position The multi-scale feature vector of the target to be identified; a judgment module, configured to set a target decision model, calculate a target recognition value of the target to be identified based on the multi-scale feature vector and in combination with environmental factors, and determine whether the target to be identified is a building, wherein the environmental factors include temperature, humidity, and air pressure; The target decision model includes: ， ， in, For the new SAR image Upper position The target identification value of the target to be identified, is the number of environmental factors, For the The weight of environmental factors, is the adjustment factor of the multi-scale feature mapping function, is the multi-scale feature mapping function, For the The first adjustment factor for environmental factors is For the The second adjustment factor for environmental factors is In the original SAR image Upper position The first environmental factors; When the new SAR image Upper position The target recognition value of the target to be identified Greater than the preset building threshold When , it is confirmed that the target to be identified is a building; Multi-scale feature mapping function include: in, is the number of scales, For the The first adjustment factor for each scale, For the The second adjustment factor for each scale, In the original SAR image Upper position The image quality correction function at , and generate a new SAR image , For the The third adjustment factor of the scale, For the The fourth adjustment factor of the scale, For the new SAR image Upper position The height gradient at For the The fifth adjustment factor of the scale, For the The sixth adjustment factor of the scale, In the original SAR image Upper position The temperature at is the standard temperature, In the original SAR image Upper position The humidity of the The standard humidity.
6. The target feature extraction and classification recognition system in SAR images according to claim 5, characterized in that: In the original SAR image Upper position Image quality correction function at include: in, In the original SAR image Upper position The terrain influence function at is used to describe the influence of terrain on target feature extraction. is the adjustment factor of the noise source, is the center coordinate of the noise source, is the standard deviation of the noise intensity.