Method and device for landmark matching in remote sensing images

Abstract

The present application discloses a method and device for landmark matching in remote sensing images. The method includes: determining the landmark area in the coastline raster image, the image gradient of the remote sensing image acquired by the geostationary orbit remote sensing satellite, and the cloud mask data; determining the similarity measurement function value based on the preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image, and the image gradient of the remote sensing image of the matching area; using the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, using the similarity measurement function as the fitness function of the particle, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measurement function value; based on the parameter value of the target parameter, resampling the coastline raster image data of the landmark area to achieve landmark matching.

Description

Technical Field

[0001] The present application relates to the field of satellite technology, and in particular to a method and device for landmark matching in remote sensing images. Background Art

[0002] The new generation of geostationary remote sensing satellites use three-axis stabilization to image the earth. Due to the influence of external factors such as orbit, attitude, thermal deformation, and internal geometric distortion factors of the remote sensor, the direction of the optical axis to the earth will change greatly within a day without control, resulting in large positioning errors in the geographic positioning of remote sensing images obtained by geostationary remote sensing satellites. In order to better eliminate the pointing deviation and system error caused by the above influences, locating landmarks in remote sensing images has become a common method.

[0003] Existing landmark matching methods mainly locate landmarks in remote sensing images based on grayscale or features. Grayscale-based landmark matching methods mainly use one-dimensional or two-dimensional sliding templates in the spatial domain for image matching, but the amount of calculation is large and the matching speed is slow. In addition, since the sea and land images with landmarks and remote sensing images belong to two types of data with different attributes, the accuracy and reliability of the landmarks located in the remote sensing images are not high; feature-based landmark matching methods extract significant features from the original image as matching primitives for feature matching, such as corner features, high curvature point features, etc. However, due to the occlusion of various factors, good matching cannot be achieved for some special areas, such as islands and rivers, and the accuracy and reliability are low. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method and device for landmark matching in remote sensing images, so as to improve the accuracy and reliability of landmark matching in remote sensing images.

[0005] In order to achieve the above objectives, the present application embodiment adopts the following technical solutions:

[0006] In a first aspect, an embodiment of the present application provides a method for landmark matching in a remote sensing image, comprising:

[0007] Determine landmark areas in coastline raster images, image gradients and cloud mask data of remote sensing images acquired by geostationary remote sensing satellites;

[0008] Determine a similarity measurement function value based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image, and the image gradient of the remote sensing image, wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area;

[0009] Using the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, using the similarity measure function as the fitness function of the particle, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value;

[0010] Based on the parameter value of the target parameter, the coastline raster image data of the landmark area is resampled to achieve landmark matching.

[0011] In a second aspect, an embodiment of the present application provides a device for matching landmarks in a remote sensing image, comprising:

[0012] The first determination module is used to determine the landmark area in the coastline raster image, the image gradient and cloud mask data of the remote sensing image acquired by the geostationary orbit remote sensing satellite;

[0013] A second determination module is used to determine a similarity measurement function value based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area;

[0014] A third determination module is used to use the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, use the similarity measure function as the fitness function of the particle, and determine the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value;

[0015] The landmark matching module is used to resample the coastline raster image data of the landmark area based on the parameter value of the target parameter to achieve landmark matching.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0017] processor;

[0018] a memory for storing instructions executable by the processor;

[0019] The processor is configured to execute the instructions to implement the method as described in the first aspect.

[0020] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the method described in the first aspect.

[0021] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0022] Considering that remote sensing images and coastline raster images belong to two types of data with different attributes, there are certain deviations in the correspondence between the image of the matching area in the remote sensing image and the coastline raster image of the area determined based on the existing grayscale correlation matching method and feature matching method, which in turn affects the accuracy and reliability of landmark matching. To this end, a preset geometric transformation function is used to represent the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching position corresponding to the landmark area in the remote sensing image, and a similarity measurement function is used to represent the similarity between the coastline raster image of the landmark area and the image of the corresponding actual area in the remote sensing image. Therefore, the purpose of landmark matching can be regarded as determining the value of the target parameter of the preset geometric transformation function when the similarity between the image of the landmark area in the coastline raster image and the image of the corresponding matching area in the remote sensing image is the largest. Thus, the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the corresponding matching position of the landmark area in the remote sensing image is obtained, and then the corresponding relationship and the coordinate position of the landmark area in the coastline raster image can be used to locate the landmark area in the remote sensing image to achieve landmark matching; further, considering that the similarity measure function is a non-convex function and there may be multiple extreme values, a heuristic swarm optimization algorithm can be adopted for this purpose, specifically a target particle swarm optimization algorithm can be adopted, by taking the target parameter of the preset geometric transformation function as the particle spatial position in the target particle swarm optimization algorithm, and taking the similarity measure function as the fitness function, based on the target particle swarm optimization algorithm and the similarity measure function, the parameter value of the target parameter of the preset geometric transformation function can be quickly and accurately determined, which is beneficial to improve the accuracy and reliability of landmark matching of candidates in remote sensing images. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] Figure 1 A flowchart of a method for landmark matching in a remote sensing image provided for one embodiment of the present application;

[0025] Figure 2 A flowchart of a method for landmark matching in a remote sensing image provided for another embodiment of the present application;

[0026] Figure 3 A schematic flow chart of a method for determining a parameter value of a target parameter provided in one embodiment of the present application;

[0027] Figure 4 A schematic structural diagram of a device for matching landmarks in a remote sensing image provided by an embodiment of the present application;

[0028] Figure 5 A schematic diagram of the structure of an electronic device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0030] The terms "first", "second", etc. in this application and the claims are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the application can be implemented in an order other than those illustrated or described here. In addition, "and / or" in this specification and the claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.

[0031] It should be understood that the acoustic model training method and speech synthesis method provided in the embodiments of the present application can be executed by an electronic device or software installed in an electronic device, specifically by a terminal device or a server device.

[0032] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0033] Please refer to Figure 1 , a method for landmark matching in a remote sensing image is provided in one embodiment of the present application, and the method may include the following steps:

[0034] S102, determining landmark areas in the coastline raster image, image gradients of remote sensing images acquired by remote sensing satellites, and cloud mask data.

[0035] Since the grayscale of pixels in remote sensing images changes significantly and the gradient value is large at the boundary between land and water, landmarks can be located in remote sensing images based on the landmark areas in the coastline raster image and the image gradient of the remote sensing image to achieve landmark matching.

[0036] In an optional implementation, determining the landmark area in the coastline raster image includes: determining the landmark area in the coastline raster image based on preset nominal grid data and the coordinate position of the matching area in the remote sensing image. More specifically, based on the coordinate position of the matching area in the remote sensing image and the preset nominal grid data, the coastline image data of the matching area can be determined, and the coastline image data of the matching area can be determined as the landmark area.

[0037] In order to ensure the accuracy of the landmark area in the coastline raster image, thereby improving the accuracy and reliability of landmark positioning in remote sensing images, in another optional implementation method, such as Figure 2 As shown, determining the landmark area in the coastline raster image can be specifically implemented as follows: based on the coastline raster image and preset nominal grid data, determining the rasterized sea-land boundary mask data; further, based on the rasterized sea-land boundary mask data, determining the landmark area in the coastline raster image.

[0038] For example, each grid unit in the nominal grid data can be taken as a whole, and a spatial index can be established according to the coordinates of the four corner points of each grid unit, and the vector node corresponding to the coastline grid image can be determined. The grid unit where the vector node is located can be determined through the spatial index, and then the pixel corresponding to the grid unit where the vector node is located is set to 1, thereby obtaining the rasterized land-sea boundary mask data. Of course, it should be understood that the land-sea boundary mask data can also be obtained according to other methods in the art, and the embodiments of the present application are not limited to this.

[0039] After obtaining the rasterized sea-land boundary mask data, optionally, the coastline image data of the matching area can be determined based on the landmark area and the matching area size of the coastline raster image, and the coastline image data of the matching area can be determined as the landmark area. In the embodiment of the present application, the image gradient of the remote sensing image can be calculated according to a preset gradient operator, such as a Sobel gradient operator, which is not limited in the embodiment of the present application.

[0040] In another embodiment of the present application, in order to enhance the contrast of remote sensing images and strengthen the interpretation and recognition capabilities of remote sensing images, thereby facilitating the improvement of the accuracy and reliability of landmark matching, such as Figure 2As shown, before the above S102, the method implemented in this application further includes: enhancing the remote sensing image. In practical applications, the enhancement of the remote sensing image can be implemented by various image enhancement methods in the art, and this embodiment of the application does not limit this.

[0041] In practical applications, the coastline raster images and remote sensing images involved in the embodiments of the present application can be generated based on preset nominal grid data.

[0042] S104, determining a similarity measurement function value based on a preset geometric transformation function, cloud mask data of the remote sensing image, landmark areas in the coastline raster image, and image gradients of the remote sensing image.

[0043] In the embodiment of the present application, the preset geometric transformation function is used to represent the coordinate position correspondence between the landmark area and the matching area corresponding to it in the remote sensing image, or in other words, the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image. Optionally, in order to accurately represent the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching area corresponding to the landmark area in the remote sensing image, the preset geometric transformation function may include an affine transformation function, as shown in the following formula (1):

[0044]

[0045] Among them, (i', j') represents the coordinate position of the landmark area in the coastline raster image; (i, j) represents the coordinate position of the matching area corresponding to the landmark area in the remote sensing image; a0 and b0 represent the parameters used to control the translation of the landmark area, and a1, a2, b1 and b2 represent the parameters used to control the rotation of the landmark area.

[0046] In an embodiment of the present application, a similarity measurement function is used to represent the similarity between the matching area corresponding to the landmark area in the remote sensing image and the actual area, or in other words, the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image and the coastline raster image data of the landmark area.

[0047] In an optional implementation, considering that the grayscale of pixels at the boundary between land and water in remote sensing images changes significantly and has a large gradient value, and also considering that factors such as deformation and cloud in remote sensing images will affect the accuracy and reliability of landmark positioning in remote sensing images, for this reason, Figure 1 and Figure 2As shown, in the above S102, determining the cloud mask data of the remote sensing image acquired by the geostationary orbit remote sensing satellite includes: performing cloud detection on the remote sensing image, and obtaining the cloud mask data of the remote sensing image based on the cloud detection result. Accordingly, the above S104 can be specifically implemented as follows: determining the convolution sum between the image gradient corresponding to the matching area in the remote sensing image, the cloud mask data of the matching area in the remote sensing image, and the coastline raster image data of the landmark area as the similarity measurement function value between the remote sensing image data of the matching area and the coastline raster image collection of the landmark area.

[0048] For example, the similarity measurement function is shown in the following formula (2):

[0049]

[0050] Among them, R represents the similarity measurement function value; D(i,j) represents the landmark area; (i,j) represents the coordinate position of the landmark area in the coastline raster image; F represents the preset geometric transformation function; (i',j') represents the coordinate position of the matching area corresponding to the landmark area in the remote sensing image; M(F(i,j)) represents the cloud mask data corresponding to the coordinate position (i',j') in the remote sensing image; Int(g) represents the rounding operation; Grad(i',j') represents the image gradient of the matching area corresponding to the landmark area in the remote sensing image; × represents the convolution operation.

[0051] It should be noted that the cloud mask data of the remote sensing image represents the mask image data generated after cloud detection of the remote sensing image, which can shield the image blocked by clouds in landmark matching. In practical applications, various detection technologies in the field can be used to detect clouds in remote sensing images, such as a single-channel cloud detection algorithm with a minimum cross entropy criterion, etc., which is not limited in the embodiments of the present application.

[0052] S106, using the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, using the similarity measurement function as the fitness function of the particle, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measurement function value.

[0053] Considering that remote sensing images and coastline raster images belong to two types of data with different attributes, the correspondence between the image of the matching area in the remote sensing image and the coastline raster image of the area determined based on the existing grayscale correlation matching method and feature matching method has certain deviations, so the accuracy and reliability are not high. The preset geometric transformation function can represent the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching position corresponding to the landmark area in the remote sensing image. The similarity measurement function can represent the similarity between the coastline raster image of the landmark area and the image of the corresponding actual area in the remote sensing image, and thus can reflect the accuracy of the landmark area located in the remote sensing image. Therefore, the purpose of landmark matching is actually to determine the value of the parameters of the preset geometric transformation function (such as the parameters a0~a2 and b0~b2 in formula (1)) when the similarity between the matching area corresponding to the landmark area in the coastline raster image and the actual area in the remote sensing image is the largest, so as to obtain the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching position corresponding to the landmark area in the remote sensing image. Furthermore, considering that the similarity measure function is a non-convex function and may have multiple extreme values, a heuristic swarm optimization algorithm can be used for this purpose, specifically a target particle swarm optimization algorithm can be used, by taking the target parameter of the preset geometric transformation function as the particle spatial position in the target particle swarm optimization algorithm, with the goal of maximizing the similarity between the matching area corresponding to the landmark area in the remote sensing image and the actual area in the coastline raster image, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function. Among them, the target parameter of the preset geometric transformation function refers to the unknown parameter in the preset geometric transformation function, for example, the target parameter of the preset geometric transformation function shown in the above formula (1) includes parameters a0~a2 and b0~b2.

[0054] In an optional implementation, in order to ensure the accuracy of the parameter values ​​of the obtained target parameters so as to subsequently improve the accuracy and reliability of landmark matching in remote sensing images, such as Figure 3 As shown, the above S106 includes:

[0055] S161, randomly generate a specified number of particle groups and initialize the spatial position and velocity of each particle in the particle group.

[0056] Specifically, the number of particles in the particle group and the initial value of the spatial position and velocity of each particle can be set according to actual needs, and the present application embodiment does not limit this. In practical applications, in order to prevent particles from exceeding the search space, it is necessary to limit the velocity of the particles, that is, v i,j ∈[-V max , V max ], where v i,jIndicates the speed of the particle.

[0057] Next, the similarity measurement function can be used as the fitness function of the particle, and the following steps S162 to S164 are repeatedly performed until the preset optimization stop condition is met. The preset optimization stop condition can be set according to actual needs, for example, the preset optimization stop condition can include that the number of iterations reaches a preset number threshold or the difference between the optimal fitness value after the previous iteration and the optimal fitness value after this iteration is less than a preset difference, etc.

[0058] S162, based on the current spatial position and current speed of each particle in the particle group, determine the first spatial position corresponding to the optimal fitness value searched by each particle in the particle group and the second spatial position corresponding to the optimal fitness value searched by the particle group.

[0059] Among them, fitness values ​​refer to the value of the similarity measurement function, that is, the similarity between the coastline grid image of the landmark area and the image of the corresponding actual area in the remote sensing image. In the embodiment of the present application, the first spatial position corresponding to the optimal fitness value searched by the particle refers to the spatial position of the particle when the similarity between the matching area corresponding to the landmark area in the coastline grid image in the remote sensing image and the actual area is maximized; the second spatial position corresponding to the optimal fitness value searched by the particle group refers to the spatial position of the particle in the particle group when the similarity between the matching area corresponding to the landmark area in the coastline grid image in the remote sensing image and the actual area is maximized.

[0060] It should be noted that the searching of the optimal fitness value by each particle in the particle group and the searching of the optimal fitness value by the particle group can be implemented in various ways, which will not be described in detail in the embodiments of the present application.

[0061] S163, updating the speed of each particle in the particle group based on the current speed, the first spatial position and the second spatial position of the particle.

[0062] For example, the velocity of each particle in the particle group can be determined by the following formula (3):

[0063]

[0064] Where i represents the number of the particle, i = 1, 2, ..., n; j represents the number of the spatial dimension, j = 1, 2, ..., D, where D is equal to the number of target parameters; represents the velocity of the i-th particle in the j-th spatial dimension at the t+1th time; represents the velocity of the i-th particle in the j-th spatial dimension before the t-th time; It represents the first spatial position corresponding to the optimal fitness value searched by the i-th particle in the j-th spatial dimension at the t-th time; It represents the second spatial position corresponding to the optimal fitness value searched by the particle group before the tth time; represents the spatial position of the i-th particle in the j-th spatial dimension at the t-th time; c1 and c2 represent acceleration factors; w represents the inertia weight, which determines the influence of the particle's historical velocity on the current velocity; r1 and r2 are random numbers in (0,1).

[0065] S164, updating the spatial position of each particle in the particle group based on the current spatial position and updated speed of each particle in the particle group.

[0066] For example, the spatial position of each particle in the particle group can be determined by the following formula (4):

[0067]

[0068] in, represents the spatial position of the i-th particle in the j-th spatial dimension at the t+1th time; represents the spatial position of the i-th particle in the j-th spatial dimension at the t-th time; represents the velocity of the i-th particle in the j-th spatial dimension at the t+1-th time.

[0069] The embodiment of the present application shows a specific implementation of the above step S106. Of course, it should be understood that step S106 can also be implemented in other ways, and the embodiment of the present application does not limit this.

[0070] S108, based on the calculated parameter value of the target parameter, resample the coastline raster image data of the landmark area to achieve landmark matching.

[0071] For example, based on the preset geometric transformation function and the parameter values ​​of the target parameters, the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching area corresponding to the landmark area in the remote sensing image can be determined; further, based on the coordinate position of the landmark area in the coastline raster image and the correspondence, the coordinate position of the matching area corresponding to the landmark area in the remote sensing image can be obtained, thereby realizing landmark matching.

[0072] In another embodiment of the present application, in order to improve the efficiency of landmark matching, after determining the landmark area in the coastline raster image based on the rasterized land-sea boundary mask data through the above S102, before the above S108, the method of the embodiment of the present application may also include: determining the rasterized land-sea boundary mask data based on the coastline raster image and the preset nominal grid data; performing overall pre-matching based on the image gradient of the remote sensing image and the rasterized land-sea boundary mask data to obtain the candidate matching area corresponding to the landmark area in the remote sensing image, and determining the search area in the remote sensing image based on the candidate matching area. Accordingly, the above S108 can be specifically implemented as: locating the landmark area in the search area based on the coordinate position of the landmark area, the preset geometric transformation function and the parameter value of the target parameter.

[0073] For example, the remote sensing image and the coastline raster image of the matching area can be roughly aligned with the rasterized sea-land boundary mask data under the projection of the preset nominal grid. However, there will still be deviations between the remote sensing image and the coastline raster image of the candidate matching area. In order to achieve better registration of the two, the overall landmark matching algorithm based on particle swarm optimization can be used to solve the specified parameter values ​​and resample the coastline raster image to achieve registration of the two data.

[0074] The method for landmark matching in a remote sensing image provided in an embodiment of the present application takes into account that remote sensing images and coastline raster images belong to two types of data with different attributes. Based on existing grayscale correlation matching methods and feature matching methods, there are certain deviations in the correspondence between the image of the matching area in the remote sensing image and the coastline raster image of the area, thereby affecting the positioning accuracy and reliability. To this end, a preset geometric transformation function is used to represent the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching position corresponding to the landmark area in the remote sensing image, and a similarity measurement function is used to represent the similarity between the coastline raster image of the landmark area and the image of the corresponding actual area in the remote sensing image. Therefore, the purpose of landmark matching can be regarded as determining the target parameter of the preset geometric transformation function when the similarity between the coastline raster image in the landmark area and the remote sensing image in the matching area is the largest. value, thereby obtaining the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the corresponding matching position of the landmark area in the remote sensing image, realizing landmark matching, and then using the correspondence and the coordinate position of the landmark area in the coastline raster image, the landmark area can be located in the remote sensing image; further, considering that the similarity measure function is a non-convex function and there may be multiple extreme values, for this reason, a heuristic swarm optimization algorithm can be adopted, specifically a target particle swarm optimization algorithm can be adopted, by taking the target parameter of the preset geometric transformation function as the particle spatial position in the target particle swarm optimization algorithm, and taking the similarity measure function as the fitness function of the particle swarm, based on the target particle swarm optimization algorithm and the similarity measure function, the parameter value of the target parameter of the preset geometric transformation function can be quickly and accurately determined, which is beneficial to improve the accuracy and reliability of the candidate landmarks located in the remote sensing image.

[0075] In addition, Figure 1 Corresponding to the method for landmark matching in remote sensing images shown in FIG. , the present application embodiment also provides a device for landmark matching in remote sensing images. Figure 4 , is a schematic structural diagram of a device 400 for matching landmarks in a remote sensing image provided by an embodiment of the present application, the device 400 comprising:

[0076] The first determination module 410 is used to determine the landmark area in the coastline raster image, the image gradient and cloud mask data of the remote sensing image acquired by the geostationary orbit remote sensing satellite;

[0077] A second determination module 420 is used to determine a similarity measurement function value based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area;

[0078] A third determination module 430 is used to use the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, use the similarity measure function as the fitness function of the particle, and determine the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value;

[0079] The landmark matching module 440 is used to resample the coastline raster image data of the landmark area based on the parameter value of the target parameter to achieve landmark matching.

[0080] The device for landmark matching in a remote sensing image provided in an embodiment of the present application takes into account that the remote sensing image and the coastline raster image belong to two types of data with different attributes. Based on the existing grayscale correlation matching method and feature matching method, the correspondence between the image of the matching area in the remote sensing image and the coastline raster image of the area is determined to have a certain deviation, which in turn affects the positioning accuracy and reliability. To this end, a preset geometric transformation function is used to represent the correspondence between the coordinate position of the landmark area in the coastline raster image and the coordinate position of the matching position corresponding to the landmark area in the remote sensing image, and a similarity measurement function is used to represent the similarity between the coastline raster image of the landmark area and the image of the corresponding actual area in the remote sensing image. Therefore, the purpose of landmark matching can be regarded as determining the value of the target parameter of the preset geometric transformation function when the similarity between the matching area corresponding to the landmark area in the coastline raster image and the actual area in the remote sensing image is the largest, thereby obtaining the coastline. The corresponding relationship between the coordinate position of the landmark area in the raster image and the coordinate position of the corresponding matching position of the landmark area in the remote sensing image is obtained, and then the corresponding relationship and the coordinate position of the landmark area in the coastline raster image are used to locate the landmark area in the remote sensing image to achieve landmark matching; further, considering that the similarity measure function is a non-convex function and there may be multiple extreme values, a heuristic swarm optimization algorithm can be used for this purpose, specifically a target particle swarm optimization algorithm can be used, by taking the target parameter of the preset geometric transformation function as the particle spatial position in the target particle swarm optimization algorithm, with the goal of maximizing the similarity between the matching area corresponding to the landmark area in the coastline raster image in the remote sensing image and the actual area, based on the target particle swarm optimization algorithm and the similarity measure function, the parameter value of the target parameter of the preset geometric transformation function can be quickly and accurately determined, which is beneficial to improve the accuracy and reliability of the candidate landmark positioning in the remote sensing image.

[0081] Optionally, the third determining module includes:

[0082] An initialization submodule, used to randomly generate a specified number of particle groups and initialize the spatial position and velocity of each particle in the particle group;

[0083] The optimization submodule is used to use the similarity measurement function as the fitness function of the particle and repeatedly perform the following iterative operations until the preset optimization stop condition is met:

[0084] Based on the current spatial position and current speed of each particle in the particle group, determining a first spatial position corresponding to the optimal fitness value searched by each particle in the particle group and a second spatial position corresponding to the optimal fitness value searched by the particle group;

[0085] Based on the current velocity of the particle, the first spatial position, and the second spatial position, updating the velocity of each particle in the particle group;

[0086] The spatial position of each particle in the particle group is updated based on the current spatial position and the updated speed of each particle in the particle group.

[0087] Optionally, the first determining module determines the cloud mask data of the remote sensing image acquired by the geostationary orbit remote sensing satellite, including:

[0088] Performing cloud detection on the remote sensing image, and obtaining cloud mask data of the remote sensing image based on the cloud detection result;

[0089] The third determination module is used to determine the convolution sum of the image gradient corresponding to the matching area in the remote sensing image, the cloud mask data of the matching area in the remote sensing image and the coastline raster image data of the landmark area as the similarity measure function value between the remote sensing image data of the matching area and the coastline raster image data of the landmark area.

[0090] Optionally, the first determining module determines the landmark area in the coastline raster image, including:

[0091] Based on the preset nominal grid data and the coordinate position of the matching area in the remote sensing image, the landmark area in the coastline grid image is determined.

[0092] Optionally, the first determination module determines the landmark area in the coastline grid image based on preset nominal grid data and the coordinate position of the matching area in the remote sensing image, including:

[0093] Determining coastline image data of the matching area based on the coordinate position of the matching area in the remote sensing image and the nominal grid data;

[0094] The coastline image data of the matching area is determined as the landmark area.

[0095] Optionally, the device further comprises:

[0096] A pre-matching module is used to perform overall pre-matching based on the image gradient of the remote sensing image and the rasterized sea-land boundary mask data before the landmark matching module resamples the coastline raster image based on the parameter value of the target parameter to obtain a candidate matching area corresponding to the landmark area in the remote sensing image;

[0097] A search area determination module, used to determine a search area in the remote sensing image based on the candidate matching area;

[0098] The landmark matching module is used to locate the landmark area in the search area based on the coordinate position of the landmark area, the preset geometric transformation function and the parameter value of the target parameter.

[0099] Optionally, the preset geometric transformation function includes an affine transformation function.

[0100] Obviously, the device for landmark matching in remote sensing images provided in the embodiment of the present application can be used as the above Figure 1 The execution body of the method for matching landmarks in remote sensing images is shown, so that the device for matching landmarks in remote sensing images can be implemented in Figure 1 Since the principle is the same, it will not be repeated here.

[0101] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0102] Figure 5 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 5 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0103] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0104] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0105] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a device for matching landmarks in remote sensing images at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0106] Determine landmark areas in coastline raster images, image gradients and cloud mask data of remote sensing images acquired by geostationary remote sensing satellites;

[0107] Determine a similarity measurement function value based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area;

[0108] Using the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, using the similarity measure function as the fitness function of the particle, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value;

[0109] Based on the parameter value of the target parameter, the coastline raster image data of the landmark area is resampled to achieve landmark matching.

[0110] The above application Figure 1The method performed by the device for matching landmarks in remote sensing images disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0111] The electronic device may also perform Figure 1 Method and device for implementing landmark matching in remote sensing images Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0112] Of course, in addition to software implementation methods, the electronic device of the present application does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0113] The embodiment of the present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by a portable electronic device including a plurality of application programs, enable the portable electronic device to execute Figure 1 The method of the embodiment shown is specifically used to perform the following operations:

[0114] Determine landmark areas in coastline raster images, image gradients and cloud mask data of remote sensing images acquired by geostationary remote sensing satellites;

[0115] Determine a similarity measurement function value based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area;

[0116] Using the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, using the similarity measure function as the fitness function of the particle, and determining the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value;

[0117] Based on the parameter value of the target parameter, the coastline raster image data of the landmark area is resampled to achieve landmark matching.

[0118] In short, the above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0119] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0120] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0121] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0122] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

Claims

1. A method for landmark matching in remote sensing images, characterized in that: include: Determining landmark areas in a coastline raster image, image gradients of a remote sensing image acquired by a geostationary remote sensing satellite, and cloud mask data, including: performing cloud detection on the remote sensing image, and obtaining cloud mask data of the remote sensing image based on the cloud detection result; Based on a preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, a similarity measurement function value is determined, including: determining the convolution sum between the image gradient corresponding to the matching area in the remote sensing image and the cloud mask data of the matching area in the remote sensing image as the similarity measurement function value between the remote sensing image data of the matching area and the coastline raster image data of the landmark area; wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area; The target parameter of the preset geometric transformation function is used as the spatial position of the particle in the target particle swarm optimization algorithm, and the similarity measure function is used as the fitness function of the particle. Based on the target particle swarm optimization algorithm and the similarity measure function value, the parameter value of the target parameter is determined; wherein the target parameter of the preset geometric transformation function refers to the unknown parameter in the preset geometric transformation function; Based on the parameter value of the target parameter, the coastline raster image data of the landmark area is resampled to achieve landmark matching.

2. The method according to claim 1, characterized in that: The target parameter of the preset geometric transformation function is used as the spatial position of the particle in the target particle swarm optimization algorithm, the similarity measurement function is used as the fitness function of the particle, and the parameter value of the target parameter is determined based on the target particle swarm optimization algorithm and the similarity measurement function value, including: Randomly generate a specified number of particle groups and initialize the spatial position and velocity of each particle in the particle group; The similarity measurement function is used as the fitness function of the particle, and the following iterative operation is repeatedly performed until the preset optimization stop condition is met: Based on the current spatial position and current speed of each particle in the particle group, determining a first spatial position corresponding to the optimal fitness value searched by each particle in the particle group and a second spatial position corresponding to the optimal fitness value searched by the particle group; Based on the current velocity of the particle, the first spatial position, and the second spatial position, updating the velocity of each particle in the particle group; The spatial position of each particle in the particle group is updated based on the current spatial position and the updated speed of each particle in the particle group.

3. The method according to claim 1, characterized in that The step of determining a landmark area in a coastline raster image comprises: Based on the preset nominal grid data and the coordinate position of the matching area in the remote sensing image, the landmark area in the coastline grid image is determined.

4. The method according to claim 3, characterized in that The step of determining the landmark area in the coastline grid image based on the preset nominal grid data and the coordinate position of the matching area in the remote sensing image comprises: Determining coastline image data of the matching area based on the coordinate position of the matching area in the remote sensing image and the nominal grid data; The coastline image data of the matching area is determined as the landmark area.

5. The method according to claim 1, characterized in that Before resampling the coastline raster image based on the parameter value of the target parameter, the method further includes: Performing overall pre-matching based on the image gradient of the remote sensing image and the rasterized sea-land boundary mask data to obtain a candidate matching area corresponding to the landmark area in the remote sensing image; Based on the candidate matching area, determining a search area in the remote sensing image; The step of resampling the coastline raster image based on the parameter value of the target parameter to achieve landmark matching includes: The landmark area is located in the search area based on the coordinate position of the landmark area, the preset geometric transformation function and the parameter value of the target parameter.

6. The method according to any one of claims 1 to 5, characterized in that The preset geometric transformation function includes an affine transformation function.

7. A remote sensing image landmark matching device, characterized in that: include: The first determination module is used to determine the landmark area in the coastline raster image, the image gradient and cloud mask data of the remote sensing image acquired by the geostationary orbit remote sensing satellite, including: performing cloud detection on the remote sensing image, and obtaining the cloud mask data of the remote sensing image based on the cloud detection result; The second determination module is used to determine the similarity measurement function value based on the preset geometric transformation function, the cloud mask data, the landmark area in the coastline raster image and the image gradient of the remote sensing image, including: determining the convolution sum between the image gradient corresponding to the matching area in the remote sensing image and the cloud mask data of the matching area in the remote sensing image as the similarity measurement function value between the remote sensing image data of the matching area and the coastline raster image data of the landmark area; wherein the preset geometric transformation function is used to represent the coordinate position correspondence between the coastline raster image data of the landmark area and the remote sensing image data of the matching area corresponding to the landmark area in the remote sensing image, and the similarity measurement function value is used to represent the similarity between the remote sensing image data of the matching area and the coastline raster image data of the landmark area; A third determination module is used to use the target parameter of the preset geometric transformation function as the spatial position of the particle in the target particle swarm optimization algorithm, and use the similarity measure function as the fitness function of the particle, and determine the parameter value of the target parameter based on the target particle swarm optimization algorithm and the similarity measure function value; wherein the target parameter of the preset geometric transformation function refers to the unknown parameter in the preset geometric transformation function; The landmark matching module is used to resample the coastline raster image data of the landmark area based on the parameter value of the target parameter to achieve landmark matching.

8. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method as claimed in any one of claims 1 to 6.

Images