Remote sensing picture generation method and system based on SAR (Synthetic Aperture Radar) and optical image fusion

Abstract

The invention discloses a remote sensing image generation method and system based on SAR and optical image fusion, and relates to the technical field of image processing, the system comprises an image monitoring module, a data acquisition module, an edge calculation module and a central server, image data and remote sensing data are acquired through a data acquisition module, a data processing model is constructed for image analysis, the quality of a source image is preliminarily analyzed and evaluated through an edge calculation module, image texture features and image color features are sequentially analyzed, image fusion and feature correction are performed through a central server, and a remote sensing image is generated; from the perspective of SAR and optical image fusion, the advantage complementation of the stability of the SAR image under poor atmospheric conditions and the high spatial resolution of the optical image is realized, the generalization ability of the system in different application scenes is improved, the real-time monitoring requirement is met through edge calculation, and the data accuracy and the generation stability of the remote sensing image are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a method and system for generating remote sensing images based on the fusion of SAR and optical images. Background Art

[0002] Remote sensing satellites provide people with rich remote sensing image data with different spatial resolutions and different spectra. These massive remote sensing images play a huge role in the fields of military, natural disaster monitoring, surveying and mapping, aerospace, etc. Due to the different imaging principles of different sensor types and the influence of natural environment and other factors, single-sensor imaging often cannot fully present all the information in the scene.

[0003] SAR (Synthetic Aperture Rader) is a type of remote sensing data for ground observation using radar technology; optical image technology uses optical devices such as cameras and video cameras to capture light and convert it into electrical signals or digital signals for remote sensing imaging.

[0004] However, the existing remote sensing image generation technologies have problems such as incomplete remote sensing data, insufficient accuracy, and unstable image generation caused by single sensors, as well as the resulting defect of insufficient system generalization ability. For example, the spatial resolution of SAR images is easily restricted. In complex terrain areas such as mountains and hilly areas, SAR imaging may show phenomena such as shadows and overlapping images. Optical images are easily restricted by weather and lighting conditions. Under conditions such as clouds, rain, snow, night, or insufficient lighting, the image quality may be affected.

[0005] In view of the above technical defects, a solution is proposed. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems of incomplete remote sensing data, insufficient accuracy, and unstable image generation caused by single remote sensing sensors in the prior art, as well as the resulting defect of insufficient system generalization ability.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions:

[0008] A method for generating remote sensing images based on the fusion of SAR and optical images, comprising the following steps:

[0009] S1, obtaining SAR images and optical images through an image monitoring module: The image monitoring module includes an SAR radar sensor and an optical sensor;

[0010] S2, the data acquisition module obtains image data and remote sensing data through SAR images and optical images. Among them, the image data includes band parameters and imaging parameters. The band parameters include the SAR band frequency and the optical band wavelength, and the imaging parameters include the spatial resolution and the temporal resolution. Then, remote sensing data A is extracted from the SAR image. Remote sensing data A includes the pixel gray values of the SAR image, and remote sensing data B is extracted from the optical image. Remote sensing data B includes the pixel chromaticity values of the optical image;

[0011] S3, the edge computing module conducts a preliminary analysis on the image data and remote sensing data: evaluates the source image quality through a preliminary analysis of the image data, and then analyzes the image texture features and image color features in sequence through remote sensing data A and remote sensing data B;

[0012] S4, the central server performs image fusion and feature correction and generates a remote sensing picture: conducts image texture and color fusion through remote sensing data A and remote sensing data B, and combines the source image quality to perform feature correction, so as to obtain remote sensing feature data and conduct deep fusion, thereby generating and outputting a remote sensing picture.

[0013] Furthermore, the image data and remote sensing data are analyzed and processed by constructing a data processing model. The data processing model includes an image quality assessment sub-model, a remote sensing data analysis sub-model, a remote sensing feature fusion sub-model, and a remote sensing picture generation sub-model;

[0014] The image data is preliminarily analyzed by the image quality assessment sub-model to evaluate the source image quality;

[0015] The remote sensing data A and remote sensing data B are classified and processed by the remote sensing data analysis sub-model, so as to obtain the texture features of the SAR image and the color features of the optical image;

[0016] The remote sensing data A and remote sensing data B are image-fused by the remote sensing feature fusion sub-model, the texture features and color features are classified and configured, and the source image quality is combined to obtain an image correction index. Feature correction is performed through the image correction index to obtain remote sensing feature data;

[0017] The remote sensing feature data is deeply fused by the remote sensing picture generation sub-model, thereby generating and outputting a remote sensing picture.

[0018] Furthermore, the acquisition marking process of the image data and remote sensing data is as follows:

[0019] The band parameters include the SAR band frequency Fa and the optical band wavelength Wb;

[0020] The imaging parameters include spatial resolution and temporal resolution; the spatial resolution and temporal resolution of the SAR image are respectively marked as SRa and TRa; the spatial resolution and temporal resolution of the optical image are respectively marked as SRb and TRb;

[0021] The remote sensing data A includes the gray values of the pixel points of the SAR image. Any pixel point of the SAR image is marked as i, and the gray value of pixel point i is marked as Gi;

[0022] The remote sensing data B includes the chromaticity values of the pixel points of the optical image. Any pixel point of the optical image is marked as j, and the chromaticity value of pixel point j is marked as Cj.

[0023] Furthermore, the specific process of the image quality assessment sub-model is as follows:

[0024] The image data is preliminarily analyzed through the image quality assessment sub-model to evaluate the quality of the source image;

[0025] S3-101, by combining the SAR band frequency Fa, the spatial resolution SRa and the temporal resolution TRa of the SAR image, the quality assessment coefficient QUsar of the SAR image is obtained;

[0026] By setting the evaluation interval of the quality assessment coefficient QUsar and conducting interval comparison to evaluate the quality of the source image of the SAR image;

[0027] S3-102, by combining the optical band wavelength Wb, the spatial resolution SRb and the temporal resolution TRb of the optical image, the quality assessment coefficient QUopt of the optical image is obtained;

[0028] By setting the evaluation interval of the quality assessment coefficient QUopt and conducting interval comparison to evaluate the quality of the source image of the optical image.

[0029] Furthermore, the specific process of the remote sensing data analysis sub-model is as follows:

[0030] The remote sensing data A and the remote sensing data B are classified through the remote sensing data analysis sub-model;

[0031] S3-201, analyze and process the remote sensing data A to obtain the texture features of the SAR image;

[0032] The SAR image is divided into m1 local regions. The number of pixel points in any local region I is marked as N1. Through the gray values Gi of the N1 pixel points in the local region I, a gray-level co-occurrence matrix is constructed;

[0033] Mark any pixel pair in the gray-level co-occurrence matrix as (x, y), where x is the row number of the gray-level co-occurrence matrix and y is the column number of the gray-level co-occurrence matrix, representing a pixel pair with gray levels (x) and (y), and mark the probability of the pixel pair (x, y) as P(x, y);

[0034] Evaluate the uniformity, contrast, and fineness of the pixel pair through the gray-level co-occurrence matrix, and then obtain the texture feature evaluation coefficient TEX of the SAR image;

[0035] S3-202, analyze and process remote sensing data B to obtain the color characteristics of the optical image;

[0036] Divide the optical image into m2 local regions, mark the number of pixels in any local region J as N2, and construct a chromaticity co-occurrence matrix through the chromaticity values Cj of the N2 pixels in the local region J;

[0037] Mark any pixel pair in the chromaticity co-occurrence matrix as (p, q), where p is the row number of the chromaticity co-occurrence matrix and q is the column number of the chromaticity co-occurrence matrix, representing a pixel pair with chromaticity values (p) and (q), and mark the probability of the pixel pair (p, q) as P(p, q);

[0038] Evaluate the richness, contrast, and saturation of the pixel pair through the chromaticity co-occurrence matrix, and then obtain the color characteristic evaluation coefficient COL of the optical image.

[0039] Further, the specific process of the remote sensing feature fusion sub-model is as follows:

[0040] Perform image fusion on remote sensing data A and remote sensing data B through the remote sensing feature fusion sub-model;

[0041] S4-101, classify and configure the texture features and color features;

[0042] Set the risk threshold R1 of the texture feature evaluation coefficient TEX of the SAR image, evaluate the texture feature state of the SAR image through threshold comparison. When the texture feature evaluation coefficient TEX of the SAR image is lower than the risk threshold R1, it is determined that the texture feature state of the SAR image is poor, and the configuration of the remote sensing data A is corrected;

[0043] Set the risk threshold R2 of the color feature evaluation coefficient COL of the optical image, evaluate the color feature state of the optical image through threshold comparison. When the color feature evaluation coefficient COL of the optical image is lower than the risk threshold R2, it is determined that the color feature state of the optical image is poor, and the configuration of the remote sensing data B is corrected;

[0044] S4-102, obtain the image correction index and then perform feature correction;

[0045] Configure and correct remote sensing data A through the quality assessment coefficient QUsar of the SAR image and the texture feature evaluation coefficient TEX to obtain the correction coefficient of the first image ;

[0046] Configure and correct remote sensing data B through the image correction index CRopt of the optical image and the color feature evaluation coefficient COL to obtain the correction coefficient of the second image ;

[0047] S4-103, obtain remote sensing feature data through feature data correction;

[0048] Establish a feature correction model and input the source image E and the parameter Re of its pixel points;

[0049] Mark the total number of pixel points of the source image E as Ne, mark any pixel point as e, obtain the neighborhood pixel points of pixel point e, preset the number of neighborhood pixel points of any pixel point e as Nu, and mark any neighborhood pixel point as f;

[0050] Substitute the source image E into the spatial coordinate system and obtain the coordinates of each pixel point;

[0051] Obtain the spatial domain weight ws(e, f) between pixel point e and pixel point f through the distance between the coordinate De of pixel point e and the coordinate Df of neighborhood pixel point f;

[0052] Then, obtain the parameter domain weight wr(e, f) between pixel point e and pixel point f through the difference between the parameter Re of pixel point e and the parameter value Rf of neighborhood pixel point f;

[0053] Combine the spatial domain weight ws(e, f) and the parameter domain weight wr(e, f) between pixel point e and Nu neighborhood pixel points f to obtain the comprehensive weight w(e,f) of pixel point e;

[0054] Set the standard interval Qw of the comprehensive weight w(e,f). When the comprehensive weight w(e,f) of pixel point e is within the standard interval Qw, it is determined that pixel point e is normal and no processing is performed on pixel point e; when the comprehensive weight w(e,f) of pixel point e is lower than or higher than the standard interval Qw, it is determined that pixel point e is abnormal, and the parameter Re of pixel point e is corrected through the image correction coefficient;

[0055] The feature correction model outputs the corrected source image E and marks it as the feature picture H.

[0056] Furthermore, the specific process of the remote sensing picture generation sub-model is as follows:

[0057] Input the SAR image into the feature correction model, obtain the abnormal pixel points of the SAR image, and perform grayscale correction, with the image correction coefficient Specifically, select the first image correction coefficient according to the actual situation , output the corrected SAR image, and label it as the SAR feature image;

[0058] Input the optical image into the feature correction model, obtain the abnormal pixel points of the optical image, and perform chromaticity correction, with the image correction coefficient Specifically, select the second image correction coefficient according to the actual situation , output the corrected optical image, and label it as the optical feature image;

[0059] Combine the SAR feature image and the optical feature image to comprehensively generate and output a remote sensing image;

[0060] Integrate and label the grayscale data of the SAR feature image and the chromaticity data of the optical feature image as remote sensing feature data. The remote sensing feature data includes the feature parameter matrix of the remote sensing image, and the feature parameter matrix includes the remote sensing feature data vectors of all pixel points , where GiR is the corrected remote sensing grayscale parameter and CjR is the corrected remote sensing chromaticity parameter.

[0061] A remote sensing image generation system based on the fusion of SAR and optical images includes an image monitoring module, a data acquisition module, an edge computing module, and a central server. The image monitoring module, the data acquisition module, the edge computing module, and the central server are communicatively connected. Among them, the edge computing module includes an image analysis sub-module and a remote sensing analysis sub-module, and the central server includes a feature correction sub-module and a depth analysis sub-module; this system applies the above-mentioned remote sensing image generation method based on the fusion of SAR and optical images.

[0062] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows:

[0063] The present invention obtains the SAR image and the optical image through the image monitoring module, obtains the image data and remote sensing data through the data acquisition module, then constructs a data processing model for image analysis, initially analyzes and evaluates the source image quality through the edge computing module and sequentially analyzes the image texture feature and the image color feature, and then performs image fusion and feature correction through the central server and generates a remote sensing image. From the perspective of the fusion of SAR and optical images, it realizes the complementary advantages of the stability of the SAR image under poor atmospheric conditions and the high spatial resolution of the optical image, improves the generalization ability of the system in different application scenarios, and meets the real-time monitoring requirements through edge computing, ensuring the data accuracy and generation stability of the remote sensing image. Description of the Drawings

[0064] Figure 1 Shows a schematic diagram of the steps of the method flow of the present invention;

[0065] Figure 2 Shows a schematic diagram of the process of the data processing model of the present invention;

[0066] Figure 3 Shows a schematic diagram of the connection of the system modules of the present invention. Detailed implementation manners

[0067] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0068] Embodiment 1:

[0069] As Figures 1-3 shown, a method for generating remote sensing images based on the fusion of SAR and optical images includes the following steps:

[0070] S1, obtaining a SAR image and an optical image through an image monitoring module: The image monitoring module includes a SAR radar sensor and an optical sensor;

[0071] Emitting electromagnetic waves through the SAR radar sensor and receiving the echoes, and recording the SAR image reflected by the ground surface; obtaining the optical image through an optical sensor that forms images using electromagnetic waves such as visible light and infrared rays;

[0072] SAR (Synthetic Aperture Radar) image: Using the principle of microwave signal emission and reflection, a high-resolution image is formed through synthetic aperture technology. SAR can not only be taken on sunny days, but also be imaged at night or under adverse weather conditions such as cloud cover and haze. However, in complex terrain areas such as mountains and hilly areas, the imaging quality of SAR may be affected, resulting in phenomena such as shadows and ghost images, reducing the accuracy and readability of the image;

[0073] Optical image: Using electromagnetic waves such as visible light and infrared rays to form images, optical imaging depends on good atmospheric conditions, so it is limited in environments such as clouds, rain, and snow;

[0074] S2, the data acquisition module obtains image data and remote sensing data through SAR images and optical images. Among them, the image data includes band parameters and imaging parameters. The band parameters include the SAR band frequency and the optical band wavelength, and the imaging parameters include the spatial resolution and the temporal resolution. Then, remote sensing data A is extracted from the SAR image, and the remote sensing data A includes the pixel gray values of the SAR image. Remote sensing data B is extracted from the optical image, and the remote sensing data B includes the pixel chromaticity values of the optical image;

[0075] The acquisition marking process of the image data and the remote sensing data is as follows:

[0076] The band parameters include the SAR band frequency Fa and the optical band wavelength Wb;

[0077] The imaging parameters include the spatial resolution and the temporal resolution. The spatial resolution and the temporal resolution of the SAR image are respectively marked as SRa and TRa; the spatial resolution and the temporal resolution of the optical image are respectively marked as SRb and TRb;

[0078] The remote sensing data A includes the pixel gray values of the SAR image. Any pixel of the SAR image is marked as i, and the gray value of the pixel i is marked as Gi;

[0079] The remote sensing data B includes the pixel chromaticity values of the optical image. Any pixel of the optical image is marked as j, and the chromaticity value of the pixel j is marked as Cj;

[0080] ‌Gray value Gi‌: Gray refers to the depth of an image or color, which is used to represent the brightness level of the image and is usually used to represent the transitional color from black to white. It is calculated by measuring the equality of RGB values, or an existing image processing tool can be used for gray recognition of pixel points;

[0081] ‌Chromaticity value Cj‌: Chromaticity refers to the basic attribute of a color, excluding brightness, and reflects the hue and saturation of the color. The color matching function is calculated by measuring the hue and saturation of the RGB color, or an existing image processing tool can be used for chromaticity recognition of pixel points;

[0082] The image data and the remote sensing data are analyzed and processed by constructing a data processing model. The data processing model includes an image quality assessment sub-model, a remote sensing data analysis sub-model, a remote sensing feature fusion sub-model, and a remote sensing picture generation sub-model;

[0083] S3, the edge computing module conducts a preliminary analysis on the image data and the remote sensing data: the source image quality is initially analyzed and evaluated through the image data, and then the image texture feature and the image color feature are analyzed in sequence through the remote sensing data A and the remote sensing data B;

[0084] S3-1. Initially analyze the image data through the image quality assessment sub-model to evaluate the quality of the source image;

[0085] The specific process of the image quality assessment sub-model is as follows:

[0086] Initially analyze the image data through the image quality assessment sub-model to evaluate the quality of the source image;

[0087] S3-101. Combine the SAR band frequency Fa, the spatial resolution SRa, and the temporal resolution TRa of the SAR image to obtain the quality assessment coefficient QUsar of the SAR image:

[0088] ;

[0089] Among them, , and are the weight coefficients of the SAR band frequency Fa, the spatial resolution SRa, and the temporal resolution TRa of the SAR image respectively, and , and are all greater than 0; when the SAR band frequency Fa, the spatial resolution SRa, and the temporal resolution TRa of the SAR image are higher, the quality assessment coefficient QUsar of the SAR image is higher, and the quality of the source image of the SAR image evaluated is better; the higher the carrier frequency, it means that the SAR image has a higher resolution and can provide clear image details. High resolution is crucial for identifying small targets, carrying out refined management, and improving the accuracy of target recognition;

[0090] Evaluate the quality of the source image of the SAR image by setting the evaluation interval of the quality assessment coefficient QUsar and conducting interval comparison;

[0091] S3-102. Combine the optical band wavelength Wb, the spatial resolution SRb, and the temporal resolution TRb of the optical image to obtain the quality assessment coefficient QUopt of the optical image:

[0092] ;

[0093] Among them, , and are the weight coefficients of the optical band wavelength Wb, the spatial resolution SRb, and the temporal resolution TRb of the optical image respectively, and , and All are greater than 0; when the wavelength Wb in the optical band, the spatial resolution SRb, and the temporal resolution TRb of the optical image are higher, the quality evaluation coefficient QUopt of the optical image is higher, and the quality of the source image of the optical image is better; the longer the wavelength, the stronger the penetration ability, and the low-frequency bands such as the [specific band] have stronger penetration ability and are suitable for penetrating vegetation, soil or observing underground targets;

[0094] By setting the evaluation interval of the quality evaluation coefficient QUopt and performing interval comparison to evaluate the quality of the source image of the optical image;

[0095] S3-2, Classify and process remote sensing data A and remote sensing data B through the remote sensing data analysis sub-model to obtain the texture features of the SAR image and the color features of the optical image;

[0096] The specific process of the remote sensing data analysis sub-model is as follows:

[0097] Classify and process remote sensing data A and remote sensing data B through the remote sensing data analysis sub-model;

[0098] S3-201, Analyze and process remote sensing data A to obtain the texture features of the SAR image;

[0099] Divide the SAR image into m1 local regions, mark the number of pixel points in any local region I as N1, and construct a gray-level co-occurrence matrix through the gray values Gi of the N1 pixel points in local region I;

[0100] Mark any pixel point pair in the gray-level co-occurrence matrix as (x, y), where x is the row number of the gray-level co-occurrence matrix and y is the column number of the gray-level co-occurrence matrix, representing the pixel point pair with gray levels (x) and (y), and mark the probability of the pixel point pair (x, y) as P(x, y);

[0101] Obtain the texture feature evaluation coefficient TEX of the SAR image through the gray-level co-occurrence matrix:

[0102]

[0103] Among them, n1 is the number of types of pixel point pairs in the gray-level co-occurrence matrix, and by traversing the uniformity , contrast and fineness of the n1 types of pixel point pairs in the gray-level co-occurrence matrix, comprehensively evaluate the texture feature state of the SAR image. When the uniformity, contrast, and fineness of the pixel point pair are higher, the texture feature evaluation coefficient TEX of the SAR image is higher, and the evaluation of the texture feature state of the SAR image is better;

[0104] S3-202, Analyze and process remote sensing data B to obtain the color features of the optical image;

[0105] Divide the optical image into m2 local regions, mark the number of pixel points in any local region J as N2, and construct a chromaticity co-occurrence matrix through the chromaticity values Cj of the N2 pixel points in the local region J;

[0106] Mark any pixel point pair in the chromaticity co-occurrence matrix as (p, q), where p is the row number of the chromaticity co-occurrence matrix and q is the column number of the chromaticity co-occurrence matrix, representing the pixel point pair with chromaticity values (p) and (q), and mark the probability of the pixel point pair (p, q) as P(p, q);

[0107] Obtain the color feature evaluation coefficient COL of the optical image through the chromaticity co-occurrence matrix:

[0108]

[0109] where n2 is the number of types of pixel point pairs in the gray-level co-occurrence matrix, Cave refers to the average value of the chromaticity values Cj of the N2 pixel points in the local region J, and by traversing the richness 、contrast 、saturation of the n2 types of pixel point pairs in the gray-level co-occurrence matrix, the color feature state of the optical image is comprehensively evaluated. The higher the richness, contrast, and saturation of the pixel point pair, the higher the color feature evaluation coefficient COL of the optical image, and the better the evaluation of the color feature state of the optical image;

[0110] S4. The central server performs image fusion and feature correction and generates a remote sensing picture: Perform image texture and color fusion on remote sensing data A and remote sensing data B, and combine the source image quality to perform feature correction to obtain remote sensing feature data and perform deep fusion, thereby generating and outputting a remote sensing picture;

[0111] S4-1. Perform image fusion on remote sensing data A and remote sensing data B through the remote sensing feature fusion sub-model, classify and configure the texture feature and the color feature, and combine the source image quality to obtain an image correction index, and perform feature correction through the image correction index to obtain remote sensing feature data;

[0112] The specific process of the remote sensing feature fusion sub-model is as follows:

[0113] Perform image fusion on remote sensing data A and remote sensing data B through the remote sensing feature fusion sub-model;

[0114] S4-101. Classify and configure the texture feature and the color feature;

[0115] Set the risk threshold R1 for the texture feature evaluation coefficient TEX of the SAR image. Evaluate the texture feature state of the SAR image by threshold comparison. When the texture feature evaluation coefficient TEX of the SAR image is lower than the risk threshold R1, it is determined that the texture feature state of the SAR image is poor, and the configuration of remote sensing data A is corrected;

[0116] Set the risk threshold R2 for the color feature evaluation coefficient COL of the optical image. Evaluate the color feature state of the optical image by threshold comparison. When the color feature evaluation coefficient COL of the optical image is lower than the risk threshold R2, it is determined that the color feature state of the optical image is poor, and the configuration of remote sensing data B is corrected;

[0117] S4-102, Obtain the image correction index and then perform feature correction;

[0118] Configure and correct the remote sensing data A through the quality evaluation coefficient QUsar and the texture feature evaluation coefficient TEX of the SAR image to obtain the first image correction coefficient : , where is the conversion coefficient of the SAR image. The conversion coefficient refers to combining the quality evaluation coefficient QUsar and the texture feature evaluation coefficient TEX and then converting them into the first image correction coefficient is a preset constant, and the preset value of is within the interval (0,1). When the quality evaluation coefficient QUsar and the texture feature evaluation coefficient TEX of the SAR image are higher, the first image correction coefficient is lower;

[0119] Configure and correct the remote sensing data B through the image correction index CRopt and the color feature evaluation coefficient COL of the optical image to obtain the second image correction coefficient : , where is the conversion coefficient of the optical image. The conversion coefficient refers to combining the image correction index CRopt and the color feature evaluation coefficient COL and then converting them into the second image correction coefficient is a preset constant, and the preset value of is within the interval (0,1). When the image correction index CRopt and the color feature evaluation coefficient COL of the optical image are higher, the second image correction coefficient is lower;

[0120] S4-103, Obtain the remote sensing feature data through feature data correction;

[0121] Establish a feature correction model and input the source image E and the parameter Re of its pixel points;

[0122] The total number of pixels of the source image E is marked as Ne, any pixel is marked as e, and the neighboring pixels of the pixel e are obtained. The number of neighboring pixels of any pixel e is preset as Nu, and any neighboring pixel is marked as f;

[0123] Substitute the source image E into the spatial coordinate system and obtain the coordinates of each pixel;

[0124] The spatial domain weight ws(e, f) between pixel e and pixel f is obtained by the distance between the coordinate De of pixel e and the coordinate Df of neighboring pixel f: ;in, It refers to the Euclidean distance between the coordinate De of the pixel point e and the coordinate Df of the neighboring pixel point f; when the distance The higher it is, the higher the spatial domain weight ws(e, f) between pixel e and pixel f is;

[0125] Then, the parameter domain weight wr(e, f) between pixel e and pixel f is obtained by the difference between the parameter Re of pixel e and the parameter value Rf of the neighboring pixel f: ;in, It refers to the difference between the parameter Re of pixel e and the parameter value Rf of neighboring pixel f; when the difference The higher it is, the higher the parameter domain weight wr(e, f) between pixel e and pixel f is;

[0126] The comprehensive weight w(e, f) of pixel e is obtained by combining the spatial domain weight ws(e, f) between pixel e and Nu neighboring pixels f and the parameter domain weight wr(e, f):

[0127] ;

[0128] Set the standard interval Qw of the comprehensive weight w(e,f). When the comprehensive weight w(e,f) of pixel e is within the standard interval Qw, the pixel e is considered normal and no processing is done on the pixel e. When the comprehensive weight w(e,f) of pixel e is lower or higher than the standard interval Qw, the pixel e is considered abnormal, indicating that the overall difference between pixel e and its neighboring pixels is too small or too large, and the parameter Re of pixel e is corrected.

[0129] The specific correction process is to compare the depth of pixel e with the neighboring pixels:

[0130] like If it is greater than 0, it means that the parameter Re of pixel e is larger than that of its neighboring pixels. The parameter Re of pixel e is reduced and the corrected parameter of pixel e is marked as RCe-: , where is the image correction coefficient;

[0131] If is less than 0, it means that the parameter Re of the pixel point e is relatively small compared to the neighboring pixel points. Increase the parameter Re of the pixel point e, and mark the parameter of the corrected pixel point e as RCe+: , where is the image correction coefficient;

[0132] The feature correction model outputs the corrected source image E and marks it as the feature picture H;

[0133] S4-2. Through the remote sensing image generation sub-model, perform deep fusion on the remote sensing feature data to generate and output the remote sensing image;

[0134] The specific process of the remote sensing image generation sub-model is as follows:

[0135] Input the SAR image into the feature correction model, obtain the abnormal pixel points of the SAR image, and perform gray-scale correction. The image correction coefficient Specifically select the first image correction coefficient according to the actual situation , output the corrected SAR image, and mark it as the SAR feature image;

[0136] Input the optical image into the feature correction model, obtain the abnormal pixel points of the optical image, and perform chromaticity correction. The image correction coefficient Specifically select the second image correction coefficient according to the actual situation , output the corrected optical image, and mark it as the optical feature image;

[0137] Combine the SAR feature image and the optical feature image to comprehensively generate and output the remote sensing image;

[0138] Integrate and mark the gray-scale data of the SAR feature image and the chromaticity data of the optical feature image as the remote sensing feature data. The remote sensing feature data includes the feature parameter matrix of the remote sensing image, and the feature parameter matrix includes the remote sensing feature data vectors of all pixel points , where GiR is the corrected remote sensing gray-scale parameter and CjR is the corrected remote sensing chromaticity parameter.

[0139] A remote sensing image generation system based on SAR and optical image fusion, including an image monitoring module, a data acquisition module, an edge computing module, and a central server. The image monitoring module, the data acquisition module, the edge computing module, and the central server are communicatively connected. Among them, the edge computing module includes an image analysis sub-module and a remote sensing analysis sub-module, and the central server includes a feature correction sub-module and a depth analysis sub-module; this system applies the above-mentioned remote sensing image generation method based on SAR and optical image fusion;

[0140] The image monitoring module acquires SAR images and optical images through SAR radar sensors and optical sensors;

[0141] The data acquisition module acquires image data and remote sensing data through SAR images and optical images;

[0142] The edge computing module conducts a preliminary analysis on the image data and remote sensing data;

[0143] The central server performs image fusion and feature correction and generates remote sensing images;

[0144] The system analyzes and processes the image data and remote sensing data by constructing a data processing model. The data processing model includes an image quality assessment sub-model, a remote sensing data analysis sub-model, a remote sensing feature fusion sub-model, and a remote sensing image generation sub-model. Among them: the image analysis sub-module applies the image quality assessment sub-model; the remote sensing analysis sub-module applies the remote sensing data analysis sub-model; the feature correction sub-module applies the remote sensing feature fusion sub-model; the depth analysis sub-module applies the remote sensing image generation sub-model.

[0145] In summary, the present invention acquires SAR images and optical images through the image monitoring module, acquires image data and remote sensing data through the data acquisition module, then constructs a data processing model for analysis, preliminarily analyzes and evaluates the quality of the source image through the edge computing module, and sequentially analyzes the image texture features and image color features, and then performs image fusion and feature correction through the central server to generate remote sensing images. From the perspective of SAR and optical image fusion, it realizes the complementary advantages of the stability of SAR images under poor atmospheric conditions and the high spatial resolution of optical images, improves the generalization ability of the model in different application scenarios, and meets the real-time monitoring requirements through edge computing, ensuring the data accuracy and generation stability of remote sensing images.

[0146] The setting of the size of the interval and threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the number of base numbers set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0147] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0148] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes should be covered within the protection scope of the present invention.

Claims

1. A method for generating remote sensing images based on the fusion of SAR and optical images, characterized in that: Including the following steps: S1. Obtain SAR images and optical images through the image monitoring module: The image monitoring module includes an SAR radar sensor and an optical sensor; S2. The data acquisition module obtains image data and remote sensing data through the SAR images and optical images. Among them, the image data includes band parameters and imaging parameters. The band parameters include the SAR band frequency and the optical band wavelength, and the imaging parameters include the spatial resolution and the temporal resolution; Then extract remote sensing data A from the SAR image. The remote sensing data A includes the pixel gray values of the SAR image, and extract remote sensing data B from the optical image. The remote sensing data B includes the pixel chromaticity values of the optical image; S3. The edge computing module conducts a preliminary analysis on the image data and remote sensing data: Evaluate the source image quality through a preliminary analysis of the image data, and then analyze the image texture features and image color features in sequence through the remote sensing data A and remote sensing data B; S4. The central server performs image fusion and feature correction and generates remote sensing pictures: Conduct image texture and color fusion through the remote sensing data A and remote sensing data B, and combine the source image quality to perform feature correction, so as to obtain remote sensing feature data and perform deep fusion, thereby generating and outputting remote sensing pictures.

2. The method for generating remote sensing images based on SAR and optical image fusion according to claim 1, characterized in that: Analyze and process the image data and remote sensing data by constructing a data processing model. The data processing model includes an image quality assessment sub-model, a remote sensing data analysis sub-model, a remote sensing feature fusion sub-model, and a remote sensing picture generation sub-model; Preliminarily analyze the image data through the image quality assessment sub-model to evaluate the source image quality; Classify and process the remote sensing data A and remote sensing data B through the remote sensing data analysis sub-model to obtain the texture features of the SAR image and the color features of the optical image; Conduct image fusion on the remote sensing data A and remote sensing data B through the remote sensing feature fusion sub-model, classify and configure the texture features and color features, and combine the source image quality to obtain an image correction index, and perform feature correction through the image correction index to obtain remote sensing feature data; Conduct deep fusion on the remote sensing feature data through the remote sensing picture generation sub-model to generate and output remote sensing pictures.

3. The remote sensing image generation method based on SAR and optical image fusion according to claim 2, characterized in that: The acquisition marking process of the image data and remote sensing data is as follows: The band parameters include the SAR band frequency Fa and the optical band wavelength Wb; The imaging parameters include the spatial resolution and the temporal resolution; Mark the spatial resolution and temporal resolution of the SAR image as SRa and TRa respectively; Mark the spatial resolution and temporal resolution of the optical image as SRb and TRb respectively; The remote sensing data A includes the pixel gray values of the SAR image. Mark any pixel of the SAR image as i, and mark the gray value of the pixel i as Gi; The remote sensing data B includes the pixel chromaticity values of the optical image. Mark any pixel of the optical image as j, and mark the chromaticity value of the pixel j as Cj.

4. The remote sensing image generation method based on SAR and optical image fusion according to claim 3, wherein: The specific process of the image quality assessment sub-model is as follows: Preliminarily analyze the image data through the image quality assessment sub-model to evaluate the source image quality; S3-101, obtain the quality evaluation coefficient QUsar of the SAR image by combining the SAR band frequency Fa, the spatial resolution SRa and the temporal resolution TRA of the SAR image; Set the evaluation interval of the quality evaluation coefficient QUsar, and evaluate the source image quality of the SAR image through interval comparison; S3-102, obtain the quality evaluation coefficient QUopt of the optical image by combining the optical band wavelength Wb, the spatial resolution SRb and the temporal resolution TRb of the optical image; Set the evaluation interval of the quality evaluation coefficient QUopt, and evaluate the source image quality of the optical image through interval comparison.

5. The remote sensing image generation method based on SAR and optical image fusion according to claim 4, characterized in that: The specific process of the remote sensing data analysis sub-model is as follows: Classify and process the remote sensing data A and the remote sensing data B through the remote sensing data analysis sub-model; S3-201, analyze and process the remote sensing data A to obtain the texture features of the SAR image; Divide the SAR image into m1 local regions, mark the number of pixel points in any local region I as N1, and construct a gray-level co-occurrence matrix through the gray values Gi of the N1 pixel points in the local region I; Mark any pixel pair in the gray-level co-occurrence matrix as (x, y), where x is the row number of the gray-level co-occurrence matrix and y is the column number of the gray-level co-occurrence matrix, representing the pixel pair with gray levels (x) and (y), and mark the probability of the pixel pair (x, y) as P(x, y); Evaluate the uniformity, contrast and fineness of the pixel pair through the gray-level co-occurrence matrix, and then obtain the texture feature evaluation coefficient TEX of the SAR image; S3-202, analyze and process the remote sensing data B to obtain the color features of the optical image; Divide the optical image into m2 local regions, mark the number of pixel points in any local region J as N2, and construct a chromaticity co-occurrence matrix through the chromaticity values Cj of the N2 pixel points in the local region J; Mark any pixel pair in the chromaticity co-occurrence matrix as (p, q), where p is the row number of the chromaticity co-occurrence matrix and q is the column number of the chromaticity co-occurrence matrix, representing the pixel pair with chromaticity values (p) and (q), and mark the probability of the pixel pair (p, q) as P(p, q); Evaluate the richness, contrast and saturation of the pixel pair through the chromaticity co-occurrence matrix, and then obtain the color feature evaluation coefficient COL of the optical image.

6. The method for generating a remote sensing image based on the fusion of SAR and optical images according to claim 5, characterized in that: The specific process of the remote sensing feature fusion sub-model is as follows: Perform image fusion on the remote sensing data A and the remote sensing data B through the remote sensing feature fusion sub-model; S4-101, classify and configure the texture features and the color features; Set the risk threshold R1 of the texture feature evaluation coefficient TEX of the SAR image, evaluate the texture feature state of the SAR image through threshold comparison. When the texture feature evaluation coefficient TEX of the SAR image is lower than the risk threshold R1, it is determined that the texture feature state of the SAR image is poor, and the configuration of the remote sensing data A is corrected; Set the risk threshold R2 for the color feature evaluation coefficient COL of the optical image, and evaluate the color feature state of the optical image through threshold comparison. When the color feature evaluation coefficient COL of the optical image is lower than the risk threshold R2, it is determined that the color feature state of the optical image is poor, and the remote sensing data B is configured and corrected; S4-102, obtain the image correction index and then perform feature correction; Configuration correction is performed on remote sensing data A through the quality evaluation coefficient QUsar of SAR images and the texture feature evaluation coefficient TEX to obtain the correction coefficient of the first image ; Configure and correct the remote sensing data B through the image correction index CRopt of the optical image and the color feature evaluation coefficient COL to obtain the correction coefficient of the second image ; S4-103, obtain the remote sensing feature data through feature data correction; Establish a feature correction model, and input the source image E and the parameter Re of its pixel points; Mark the total number of pixel points of the source image E as Ne, mark any pixel point as e, obtain the neighboring pixel points of the pixel point e, preset the number of neighboring pixel points of any pixel point e as Nu, and mark any neighboring pixel point as f; Substitute the source image E into the spatial coordinate system and obtain the coordinates of each pixel point; Obtain the spatial domain weight ws(e, f) between the pixel point e and the pixel point f through the distance between the coordinate De of the pixel point e and the coordinate Df of the neighboring pixel point f; Then, obtain the parameter domain weight wr(e, f) between the pixel point e and the neighboring pixel point f through the difference between the parameter Re of the pixel point e and the parameter value Rf of the neighboring pixel point f; Combine the spatial domain weight ws(e, f) and the parameter domain weight wr(e, f) between the pixel point e and the Nu neighboring pixel points f to obtain the comprehensive weight w(e, f) of the pixel point e; Set the standard interval Qw of the comprehensive weight w(e, f). When the comprehensive weight w(e, f) of the pixel point e is within the standard interval Qw, it is determined that the pixel point e is normal and no processing is performed on the pixel point e; when the comprehensive weight w(e, f) of the pixel point e is lower than or higher than the standard interval Qw, it is determined that the pixel point e is abnormal, and the parameter Re of the pixel point e is corrected through the image correction coefficient; The feature correction model outputs the corrected source image E and marks it as the feature picture H.

7. The method for generating remote sensing images based on the fusion of SAR and optical images according to claim 6, wherein: The specific process of the remote sensing picture generation sub-model is as follows: Input the SAR image into the feature correction model to obtain the abnormal pixel points of the SAR image and perform gray correction, and the image correction coefficient Specifically, select the first image correction coefficient according to the actual situation , output the corrected SAR image and label it as the SAR feature image; Input the optical image into the feature correction model, obtain the abnormal pixel points of the optical image, and perform chromaticity correction and image correction coefficient Specifically, select the second image correction coefficient according to the actual situation , output the corrected optical image, and mark it as the optical feature image; Combine the SAR feature image and the optical feature image to comprehensively generate and output the remote sensing picture; Integrate the grayscale data of the SAR feature image and the chromaticity data of the optical feature image and label it as remote sensing feature data. The remote sensing feature data includes the feature parameter matrix of the remote sensing image, and the feature parameter matrix includes the remote sensing feature data vectors of all pixel points , where GiR is the corrected remote sensing grayscale parameter and CjR is the corrected remote sensing chromaticity parameter.

8. A remote sensing image generation system based on the fusion of SAR and optical images, characterized in that: It includes an image monitoring module, a data acquisition module, an edge computing module, and a central server. The image monitoring module, the data acquisition module, the edge computing module, and the central server are communicatively connected. Among them, the edge computing module includes an image analysis sub-module and a remote sensing analysis sub-module, and the central server includes a feature correction sub-module and a depth analysis sub-module; this system applies the remote sensing picture generation method based on the fusion of SAR and optical images described in any one of claims 1-7 above.

Images