Positioning method, device and equipment based on optical satellite constellation and medium

Abstract

The embodiment of the invention discloses a positioning method and device based on an optical satellite constellation, equipment and a medium, and relates to the technical field of navigation positioning, and the method comprises the steps: obtaining a to-be-positioned image collected by any optical satellite, inputting the to-be-positioned image into a trained matchable judgment model, and carrying out the matching of the to-be-positioned image, extracting a matchable area based on the to-be-positioned image through a matchable judgment model; performing matching based on the matchable region and a pre-constructed high-precision digital surface model to obtain a plane elevation control point; and performing adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculating longitude and latitude elevation information of the to-be-positioned target by combining the geometric positioning model, the high-precision digital surface model and the to-be-positioned image. Matchable area judgment is carried out through the high-precision digital surface model, the matching time of the image and the high-precision digital surface model can be shortened, the satellite processing pressure is reduced, and the positioning precision of the satellite image is improved.

Description

Technical Field

[0001] The present application relates to the technical field of navigation and positioning, and in particular, to a positioning method, device, equipment and medium based on an optical satellite constellation. Background Art

[0002] At present, in order to improve the real-time performance of remote sensing observations, low-cost and large-scale rapid networking constellations have become an important trend in the development of space remote sensing technology. Through a constellation composed of a large number of optical satellites, near-real-time observations of global hot spots can be carried out. For example, "Jilin-1" has deployed 115 on-orbit high-resolution optical satellites.

[0003] Satellite constellations can greatly improve the response ability to observation data. However, due to the long traditional "shooting - downlink - receiving - processing" link, there is still a large delay from observation to response, and it is still difficult to meet the real-time response requirements. Combining on-orbit processing technology can nearly real-time identify targets and regions of interest, reduce data redundancy, and reduce data transmission delay, thereby improving the in-orbit real-time positioning accuracy.

[0004] In order to save the cost of constellation networking, micro and nano-satellite constellations can be formed by networking micro and nano-satellites. On the one hand, although the measurement equipment cost of micro and nano-satellite constellations is relatively low, the corresponding hardware measurement accuracy is limited and the stability is poor. When performing in-orbit real-time positioning through micro and nano-satellite constellations, the accuracy of the obtained positioning results is relatively low and it is difficult to meet the actual application requirements. On the other hand, in traditional satellite photogrammetry processing, high-precision digital orthophoto maps (DOM) and digital elevation models (DEM) are usually used as control base maps to provide control points for satellite images. However, the scheme of combining DOM and DEM has storage redundancy and is not suitable for low-cost satellites with limited storage resources, and thus is not suitable for micro and nano-satellite constellations.

[0005] Therefore, there is currently a lack of a positioning method that can be applied to micro and nano-satellite constellations and can improve the in-orbit positioning accuracy. Summary of the Invention

[0006] Embodiments of the present application provide a positioning method, device, equipment and storage medium based on an optical satellite constellation to solve the defects of the above related technologies. The technical solutions are as follows: In a first aspect, embodiments of the present application provide a positioning method based on an optical satellite constellation. The method includes: Obtain the to-be-positioned image collected by any optical satellite, input the to-be-positioned image into the trained match judgment model, and extract the matchable region from the to-be-positioned image through the match judgment model; Match based on the matchable region and the pre-constructed high-precision digital surface model to obtain plane elevation control points; Perform adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-positioned target by combining the geometric positioning model, the high-precision digital surface model, and the to-be-positioned image; Wherein, the high-precision digital surface model is constructed based on the digital orthophoto map, the digital elevation model, the lidar data, and the multi-view optical remote sensing images collected by multiple optical satellites in the optical satellite constellation.

[0007] In an alternative scheme of the first aspect, the construction process of the high-precision digital surface model includes the following steps: Perform adjustment processing on the multi-view optical remote sensing images based on the digital orthophoto map, the digital elevation model, and the lidar data; Extract multiple stereo image pairs formed by any two images from the adjusted multi-view optical remote sensing images; Based on the time interval and intersection angle between the two images corresponding to each stereo pair, filter out the stereo image pairs that meet the time interval condition and the intersection angle condition; Construct the high-precision digital surface model based on all the filtered stereo image pairs.

[0008] In an alternative scheme of the first aspect, the step of the match judgment model extracting the matchable region from the to-be-positioned image includes: Divide the to-be-positioned image into multiple to-be-matched image blocks; Perform feature extraction on each of the to-be-matched image blocks respectively to obtain the corresponding gradient map, frequency map, and grayscale map; Perform feature extraction on the gradient map, frequency map, and grayscale map respectively to obtain the gradient feature map, frequency feature map, and grayscale feature map; Perform feature fusion based on the gradient feature map, frequency feature map, and grayscale feature map to obtain the fusion feature; Perform feature extraction based on the fusion feature to obtain the match discrimination information, and process the match discrimination information through the fully connected layer and the softmax layer to obtain the type of the to-be-matched image block, and determine whether it is a matchable region; Output all the matchable regions.

[0009] In an alternative solution of the first aspect, the matching based on the matchable region and the high-precision digital surface model to obtain plane elevation control points includes: Taking the to-be-matched image block corresponding to the matchable region as the reference image; Extracting homologous points of the reference image and the high-precision digital surface model through a feature matching algorithm; Based on the extracted homologous points, obtaining plane elevation control points on the high-precision digital surface model.

[0010] In an alternative solution of the first aspect, the training process of the matchable judgment model includes: Inputting the constructed training set into the matchable judgment model; wherein, the training set takes a sample remote sensing image as input and the sample types of each sample remote sensing image block obtained by dividing the sample remote sensing image as output, and the sample types include matchable types and other types; Performing block processing on the sample remote sensing image through the matchable judgment model to obtain multiple sample remote sensing image blocks, and respectively outputting estimated types based on each sample remote sensing image block; Constructing a loss function based on the difference between the estimated type of each sample remote sensing image block and the sample type; Based on the loss function, determining whether the matchable judgment model converges, outputting the parameters of the converged matchable judgment model, and obtaining the trained matchable judgment model.

[0011] In an alternative solution of the first aspect, the construction steps of the training set include: Obtaining sample remote sensing images collected by multiple optical satellites in the optical satellite constellation; Dividing the sample remote sensing image into a preset number of sample remote sensing image blocks; Respectively extracting homologous points between each sample remote sensing image block and the high-precision digital surface model through a feature matching algorithm, and determining the spatial distribution characteristics and quantity of all homologous points corresponding to each sample remote sensing image block; Screening out sample remote sensing image blocks with matching quality meeting the preset requirements according to the spatial distribution characteristics and quantity, and the types of the screened sample remote sensing image blocks are matchable types; Taking the screened sample remote sensing image blocks as input and the types of the screened sample remote sensing image blocks as output, constructing the training set.

[0012] In an alternative solution of the first aspect, after obtaining the sample remote sensing images collected by multiple optical satellites in the optical satellite constellation, it further includes: Adjust the image parameters of each of the sample remote sensing images through gamma transformation technology, and expand to obtain the extended sample remote sensing images of each of the sample remote sensing images under different illumination conditions; Perform the step of constructing a training set based on all the extended sample remote sensing images and the sample remote sensing images.

[0013] In a second aspect, an embodiment of the present application further provides a positioning method device based on an optical satellite constellation, including: A data acquisition module, configured to acquire a to-be-positioned image collected by any optical satellite, input the to-be-positioned image into a trained match judgment model, and extract a matchable area from the to-be-positioned image through the match judgment model; A data matching module, configured to perform matching based on the matchable area and a pre-constructed high-precision digital surface model to obtain plane elevation control points; A calculation module, configured to perform adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-positioned target by combining the geometric positioning model, the high-precision digital surface model, and the to-be-positioned image; Wherein, the high-precision digital surface model is constructed based on a digital orthophoto map, a digital elevation model, laser data, and multi-view optical remote sensing images collected by multiple optical satellites in the optical satellite constellation.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the method provided in the first aspect or any implementation manner of the first aspect of the embodiments of the present application is implemented.

[0015] In a fourth aspect, the present application further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method provided in the first aspect or any implementation manner of the first aspect of the embodiments of the present application is implemented.

[0016] The beneficial effects brought by the technical solutions provided in some embodiments of the present application at least include: The present application can achieve on-board positioning based on small and micro-nano satellites. By using a high-precision digital surface model as a control base map, the elevation information required for positioning can be guaranteed, and on-board high-precision positioning can also be completed with less resource occupation.

[0017] A high-precision digital surface model can be established using remote sensing data produced by an optical satellite constellation, and then on-board high-precision positioning can be completed through the high-precision digital surface model, realizing data self-sufficiency of small and micro-nano satellite constellations.

[0018] Judging the matchable area through a high-precision digital surface model can reduce the matching time between the image and the high-precision digital surface model, reduce the on-board processing pressure, and improve the positioning accuracy of satellite images. Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions in the present application or related technologies, the following will briefly introduce the drawings required for use in the embodiments or related technology descriptions. Obviously, the drawings in the following descriptions are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0020] Figure 1 It is a schematic flowchart of a positioning method based on an optical satellite constellation provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a positioning device based on an optical satellite constellation provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed Embodiments

[0021] To make the objectives, technical solutions, and advantages of the present application clearer, the following will clearly and completely describe the technical solutions in the present application with reference to the drawings in the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0022] The terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the listed steps or modules, but optionally further includes steps or modules not listed, or optionally further includes other steps or modules inherent to these processes, methods, products, or devices.

[0023] It should be noted that the terms "first" and "second" involved in the present application are only used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first" and "second" can be interchanged in a specific order or sequence when permitted. It should be understood that the objects distinguished by "first" and "second" can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those described or illustrated here.

[0024] The following will describe the present application in detail with reference to specific embodiments.

[0025] Next, in combination with Figure 1 , a positioning method based on an optical satellite constellation provided by an embodiment of the present application will be introduced. Specifically, please refer to Figure 1 . Figure 1 FIG. shows a schematic flowchart of a positioning method based on an optical satellite constellation provided by an embodiment of the present application. As Figure 1 shown, the method includes the following steps: S101, obtain a to-be-positioned image collected by any optical satellite, input the to-be-positioned image into a trained match judgment model, and extract a matchable area based on the to-be-positioned image through the match judgment model; S102, perform matching based on the matchable area and a pre-constructed high-precision digital surface model to obtain a plane elevation control point; S103, perform adjustment based on the plane elevation control point to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-positioned target in combination with the geometric positioning model, the high-precision digital surface model, and the to-be-positioned image; In some embodiments, the high-precision digital surface model (Digital Surface Model, DSM) is constructed based on a digital orthophoto map (Digital Orthophoto Map, DOM), a digital elevation model (Digital Elevation Model, DEM), laser data, and multi-view optical remote sensing images collected by multiple optical satellites in the optical satellite constellation.

[0026] It should be noted that the DSM not only includes the elevation information of the terrain, but also includes the elevation information of surface coverings (such as buildings, vegetation, etc.). The DOM is a high-resolution image with coordinates cut according to the topographic map range through geometric correction and image mosaicing of aerial or spaceborne images, such as Google Maps, Tianditu, etc. The DEM is a digital model representing the ground terrain, only including the ground height information, and not including other objects on the ground surface, such as buildings, vegetation, etc., such as SRTM (Shuttle Radar Topography Mission) DEM, Copernicus DEM, etc. The laser data can be point cloud data of the ground surface, such as icesat-2.

[0027] In some embodiments, the construction process of the high-precision digital surface model includes the following steps: The construction of a high-precision digital surface model (DSM) can be completed on ground-based devices. Multiple optical remote sensing images can be acquired by multiple satellites in an optical satellite constellation. Each image in the multiple optical remote sensing images can be a remote sensing image from a different perspective or a different satellite. The multiple optical remote sensing images from different perspectives or satellites can be preprocessed to improve the accuracy of each image and ensure the consistency of the resolution.

[0028] Based on the digital orthophoto map, the digital elevation model, and the lidar data, the multiple optical remote sensing images are adjusted to improve the positioning accuracy.

[0029] Extract any two images from the adjusted multiple optical remote sensing images to form multiple stereo image pairs.

[0030] Based on the time interval and intersection angle between the two images corresponding to each stereo pair, stereo pairs that meet the time interval condition and the intersection angle condition are selected.

[0031] Specifically, the following constraint conditions can be used to select stereo pairs: ; Among them, and are any two scenes in the multiple optical remote sensing images. E(*) is the weighted value of the two scenes as a stereo pair, time(*) is the weight of the time interval between the two scenes, and angle(*) is the weight of the intersection angle between the two scenes. The weighted value of the two weights is compared with a preset threshold. If it is greater than the preset threshold, it is determined that it can be retained as the selected stereo pair.

[0032] Based on all the selected stereo pairs, through epipolar rectification, dense matching, point cloud reconstruction, and DSM fusion preprocessing, the high-precision digital surface model can be constructed.

[0033] In some embodiments, in S101, the steps of the matchable judgment model extracting the matchable region based on the image to be located include: Divide the image to be located into multiple image blocks to be matched; Extract features from each of the image blocks to be matched to obtain corresponding gradient maps, frequency maps, and grayscale maps; Extract features from the gradient map, frequency map, and grayscale map respectively to obtain a gradient feature map, a frequency feature map, and a grayscale feature map; Perform feature fusion based on the gradient feature map, frequency feature map, and grayscale feature map to obtain a fused feature; Feature extraction is performed based on the fused features to obtain matching discrimination information. The matching discrimination information is processed through a fully connected layer and a softmax layer to obtain the type of the image block to be matched, and it is determined whether it is a matchable area. Output all matchable areas.

[0034] Specifically, after the matchable judgment model obtains the image block to be matched, feature extraction can be performed through a convolutional layer to obtain corresponding gradient maps, frequency maps, and grayscale maps. The gradient map, frequency map, and grayscale map respectively undergo feature extraction through several convolutional layers to obtain a gradient feature map, a frequency feature map, and a grayscale feature map, so as to enhance the local feature representation ability of the image. Subsequently, using a weighted fusion strategy, the outputs of different feature maps are concatenated to fully combine multi-modal information and ensure more accurate discrimination of the matching area. After completing feature fusion, the features are further processed deeply through multiple convolutional layers to extract more abstract matching discrimination information. The matching discrimination information can be understood as a high-dimensional vector representing the image features obtained through processing. Finally, the matching discrimination information is input into the fully connected layer and the Softmax layer to process and obtain the type of the image block to be matched. For example, a prediction vector composed of the probabilities of each type can be output, and the maximum value in the prediction vector is taken to determine whether the type of the image block to be matched is a matchable area.

[0035] Based on this, through a variety of feature fusions and the classification ability of the deep learning network, the best matching areas can be quickly screened out on the satellite, reducing the occurrence of non-matches, thereby improving the geometric positioning accuracy.

[0036] In some embodiments, in S102, matching is performed based on the matchable area and the high-precision digital surface model to obtain plane elevation control points, including: Taking the image block to be matched corresponding to the matchable area as the reference image; Extracting homologous points between the reference image and the high-precision digital surface model through a feature matching algorithm; Based on the extracted homologous points, plane elevation control points on the high-precision digital surface model are obtained.

[0037] In some embodiments, the training process of the matchable judgment model includes: Inputting the constructed training set into the matchable judgment model; wherein, the training set takes sample remote sensing images as input and the sample types of each sample remote sensing image block obtained by dividing the sample remote sensing images as output. The sample types include matchable types and other types; The sample remote sensing images are divided into blocks through the matchable judgment model to obtain multiple sample remote sensing image blocks, and the estimated types are output respectively based on each sample remote sensing image block; Construct a loss function based on the difference between the estimated type of each sample remote sensing image block and the sample type; Based on the loss function, determine whether the match judgment model converges, output the parameters of the converged match judgment model, and obtain the trained match judgment model.

[0038] In some embodiments, the steps for constructing the training set include: Obtain sample remote sensing images collected by multiple optical satellites in an optical satellite constellation; Divide the sample remote sensing images into a preset number of sample remote sensing image blocks; Extract corresponding points between each sample remote sensing image block and the high-precision digital surface model respectively through a feature matching algorithm, and determine the spatial distribution characteristics and quantity of all corresponding points corresponding to each sample remote sensing image block; Screen out sample remote sensing image blocks whose matching quality meets the preset requirements according to the spatial distribution characteristics and quantity, and the type of the screened sample remote sensing image blocks is the matchable type; Using the screened sample remote sensing image blocks as input and the type of the screened sample remote sensing image blocks as output, construct the training set.

[0039] In some embodiments, after obtaining the sample remote sensing images collected by multiple optical satellites in the optical satellite constellation, it further includes: Adjust the image parameters of each sample remote sensing image through gamma transformation technology to obtain extended sample remote sensing images of each sample remote sensing image under different illumination conditions. The image parameters include but are not limited to the contrast, resolution, etc. of the image, and the embodiments of the present application do not limit this.

[0040] Execute the steps for constructing the training set based on all the extended sample remote sensing images and the sample remote sensing images.

[0041] In this way, the diversity of the training set can be expanded, the generalization ability of the match judgment model in practical applications can be improved, and it can more robustly handle the matching difficulties brought by illumination changes.

[0042] The following is the device embodiment of the present application, which can be used to execute the method embodiment of the present application. For details not disclosed in the device embodiment of the present application, please refer to the method embodiment of the present application.

[0043] Next, please refer to Figure 2, which is a schematic structural diagram of a positioning device based on an optical satellite constellation provided by an exemplary embodiment of the present application. The device can be implemented as all or part of a terminal through software, hardware, or a combination of both, and can also be integrated as an independent module on a server. The positioning device based on an optical satellite constellation in the embodiments of the present application can be applied to a terminal or the cloud. The device 20 includes a data acquisition module 201, a data matching module 202, and a calculation module 203, where: The data acquisition module 201 is configured to acquire a to-be-positioned image collected by any optical satellite, input the to-be-positioned image into a trained matchable judgment model, and extract a matchable area based on the to-be-positioned image through the matchable judgment model; The data matching module 202 is configured to perform matching based on the matchable area and a pre-constructed high-precision digital surface model to obtain plane elevation control points; The calculation module 203 is configured to perform adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-positioned target by combining the geometric positioning model, the high-precision digital surface model, and the to-be-positioned image.

[0044] It should be noted that when the device 20 provided in the above embodiment executes the positioning method based on an optical satellite constellation, only the above-mentioned division of each functional module is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the embodiment of the positioning method based on an optical satellite constellation belong to the same concept, and the implementation process thereof can be seen in the method embodiment, which will not be elaborated here.

[0045] The embodiments of the present application further provide an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the method in any of the above embodiments are implemented.

[0046] Please refer to Figure 3 , which is a structural block diagram of an electronic device provided by an embodiment of the present application.

[0047] As Figure 3 shown, the electronic device 300 includes a processor 301 and a memory 302.

[0048] In the embodiments of the present application, the processor 301 is the control center of the computer system, which may be the processor of a physical machine or the processor of a virtual machine. The processor 301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array).

[0049] The processor 301 may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake state, also known as the CPU (Central Processing Unit); the coprocessor is a low-power processor used to process data in the standby state.

[0050] The memory 302 may include one or more computer-readable storage media, and the computer-readable storage media may be non-transitory. The memory 302 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In some embodiments of the present application, the non-transitory computer-readable storage media in the memory 302 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 301 to implement the method in the embodiments of the present application.

[0051] In some embodiments, the electronic device 300 further includes: a peripheral device interface 303 and at least one peripheral device 304. The processor 301, the memory 302, and the peripheral device interface 303 may be connected through a bus or signal lines. Each peripheral device 304 may be connected to the peripheral device interface 303 through a bus, signal lines, or a circuit board. Specifically, the peripheral device 304 includes: a display screen, a camera, and an audio circuit. The peripheral device interface 303 may be used to connect at least one I / O (Input / Output) related peripheral device to the processor 301 and the memory 302.

[0052] In some embodiments of the present application, the processor 301, the memory 302, and the peripheral device interface 303 are integrated on the same chip or circuit board; in some other embodiments of the present application, any one or two of the processor 301, the memory 302, and the peripheral device interface 303 may be implemented on a separate chip or circuit board. The embodiments of the present application do not make specific limitations in this regard.

[0053] The block diagram of the electronic device shown in the embodiments of the present application does not limit the electronic device 300. The electronic device 300 may include more or fewer components than those shown in the figure, combine certain components, or adopt different component arrangements.

[0054] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the method in any of the foregoing embodiments are implemented. Among them, the computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, and magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0055] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solutions, or the part that contributes to the related technologies, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disks, optical disks, etc., and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A positioning method based on an optical satellite constellation, characterized in that, Including: Obtain the to-be-located image collected by any optical satellite, input the to-be-located image into the trained match judgment model, and extract the matchable region based on the to-be-located image through the match judgment model; Match based on the matchable region and the pre-constructed high-precision digital surface model to obtain plane elevation control points; Perform adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-located target by combining the geometric positioning model, the high-precision digital surface model, and the to-be-located image; Wherein, the high-precision digital surface model is constructed based on the digital orthophoto map, the digital elevation model, the lidar data, and the multi-view optical remote sensing images collected by multiple optical satellites in the optical satellite constellation.

2. The positioning method based on an optical satellite constellation according to claim 1, wherein The construction process of the high-precision digital surface model includes the following steps: Perform adjustment processing on the multi-view optical remote sensing images based on the digital orthophoto map, the digital elevation model, and the lidar data; Extract multiple stereo image pairs formed by any two of the multi-view optical remote sensing images after the adjustment processing; Based on the time interval and intersection angle between the two images corresponding to each stereo pair, screen out the stereo pairs that meet the time interval condition and the intersection angle condition; Construct the high-precision digital surface model based on all the screened stereo pairs.

3. The positioning method based on an optical satellite constellation according to claim 2, wherein The step of the match judgment model extracting the matchable region based on the to-be-located image includes: Divide the to-be-located image into blocks to obtain multiple to-be-matched image blocks; Perform feature extraction on each of the to-be-matched image blocks respectively to obtain the corresponding gradient map, frequency map, and grayscale map; Perform feature extraction on the gradient map, frequency map, and grayscale map respectively to obtain the gradient feature map, frequency feature map, and grayscale feature map; Perform feature fusion based on the gradient feature map, frequency feature map, and grayscale feature map to obtain the fusion feature; Perform feature extraction based on the fusion feature to obtain the match discrimination information, and process the match discrimination information through the fully connected layer and the softmax layer to obtain the type of the to-be-matched image block, and determine whether it is a matchable region; Output all the matchable regions.

4. The positioning method based on an optical satellite constellation according to claim 3, wherein The matching based on the matchable region and the high-precision digital surface model to obtain the plane elevation control points includes: Use the to-be-matched image block corresponding to the matchable region as the reference image; Extract the homologous points of the reference image and the high-precision digital surface model through the feature matching algorithm; Extract the plane elevation control points on the high-precision digital surface model based on the homologous points.

5. The positioning method based on an optical satellite constellation according to claim 2, wherein, The training process of the match judgment model includes: Input the constructed training set into the match judgment model; wherein, the training set takes the sample remote sensing image as the input and the sample types of each sample remote sensing image block obtained by dividing the sample remote sensing image as the output, and the sample types include the matchable type and other types; Perform block processing on the sample remote sensing image through the match judgment model to obtain multiple sample remote sensing image blocks, and respectively output the estimated types based on each sample remote sensing image block; Construct a loss function based on the difference between the estimated type of each sample remote sensing image block and the sample type; Based on the loss function, determine whether the matchable judgment model converges, output the parameters of the converged matchable judgment model, and obtain the trained matchable judgment model.

6. The positioning method based on an optical satellite constellation according to claim 5, characterized in that, The construction steps of the training set include: Obtain sample remote sensing images collected by multiple optical satellites in the optical satellite constellation; Divide the sample remote sensing images into a preset number of sample remote sensing image blocks; Extract the corresponding points between each sample remote sensing image block and the high-precision digital surface model through a feature matching algorithm, and determine the spatial distribution characteristics and quantity of all corresponding points corresponding to each sample remote sensing image block; Screen the sample remote sensing image blocks whose matching quality meets the preset requirements according to the spatial distribution characteristics and quantity, and the type of the screened sample remote sensing image blocks is the matchable type; Construct the training set with the screened sample remote sensing image blocks as the input and the type of the screened sample remote sensing image blocks as the output.

7. A positioning method based on an optical satellite constellation according to claim 6, characterized in that, After obtaining the sample remote sensing images collected by multiple optical satellites in the optical satellite constellation, it further includes: Adjust the image parameters of each sample remote sensing image through gamma transformation technology, and expand to obtain the extended sample remote sensing images of each sample remote sensing image under different illumination conditions; Execute the steps of constructing the training set based on all the extended sample remote sensing images and the sample remote sensing images.

8. A positioning device based on an optical satellite constellation, characterized in that, It includes: A data acquisition module, configured to acquire a to-be-located image collected by any optical satellite, input the to-be-located image into the trained matchable judgment model, and extract a matchable area based on the to-be-located image through the matchable judgment model; A data matching module, configured to perform matching based on the matchable area and a pre-constructed high-precision digital surface model to obtain plane elevation control points; A calculation module, configured to perform adjustment based on the plane elevation control points to obtain a corrected geometric positioning model, and calculate the longitude, latitude, and elevation information of the to-be-located target in combination with the geometric positioning model, the high-precision digital surface model, and the to-be-located image; Wherein, the high-precision digital surface model is constructed based on a digital orthophoto map, a digital elevation model, lidar data, and multi-view optical remote sensing images collected by multiple optical satellites in the optical satellite constellation.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the program, it implements the steps of the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

Images