Method and device for evaluating morphological characteristics of Mars barrier impact crater and storage medium

Abstract

The invention discloses an evaluation method and device for morphological characteristics of a Mars barrier impact crater and a storage medium, relates to the technical field of planetary remote sensing, and mainly aims to improve the evaluation efficiency and the evaluation precision of the morphological characteristics of the Mars barrier impact crater. The method comprises the steps of obtaining position information of a target barrier impact pit in response to a morphological characteristic evaluation signal of the target barrier impact pit; based on the position information, determining a remote sensing image to which the target barrier impact crater belongs, and matching a digital elevation model homologous with the remote sensing image; based on the remote sensing image and the digital elevation model, segmenting the target barrier impact crater and a contour area of a corresponding sputtering object in the remote sensing image; determining geometrical characteristic parameters of the contour area, and evaluating the morphological characteristics of the target barrier impact crater based on the geometrical characteristic parameters. The method is suitable for the scene of evaluating the morphological characteristics of the Mars barrier impact crater.

Description

Technical Field

[0001] The present invention relates to the field of planetary remote sensing technology, and in particular to a method, device and storage medium for evaluating the morphological characteristics of a Martian rampart impact crater. Background Art

[0002] Rampart craters are widely distributed on the surface of Mars, and these craters are usually accompanied by geomorphic features related to fluids. In order to reveal the characteristics of impact events and understand the geological structure of Mars, it is necessary to evaluate the morphological characteristics of the rampart craters on Mars.

[0003] At present, the morphological characteristics of the Martian rampart crater are usually identified by human eyes. However, this manual evaluation method is greatly affected by human subjective factors, resulting in large errors in the evaluation of the morphological characteristics of the Martian rampart crater. At the same time, manual evaluation is time-consuming and labor-intensive. Summary of the invention

[0004] The present invention provides a method, device and storage medium for evaluating the morphological characteristics of a Martian barrier crater, which are mainly capable of improving the evaluation efficiency and accuracy of the morphological characteristics of a Martian barrier crater.

[0005] According to a first aspect of the present invention, there is provided a method for evaluating the morphological characteristics of a Martian rampart impact crater, comprising: In response to a morphological feature evaluation signal of a target barrier impact crater, obtaining position information of the target barrier impact crater; Based on the location information, determining the remote sensing image to which the target barrier crater belongs, and matching a digital elevation model homologous to the remote sensing image; Based on the remote sensing image and the digital elevation model, segmenting the target barrier impact crater and the contour area of ​​the corresponding sputtering in the remote sensing image; The geometric characteristic parameters of the contour area are determined, and based on the geometric characteristic parameters, the morphological characteristics of the target barrier impact crater are evaluated.

[0006] Optionally, the location information is the latitude and longitude coordinates of the target barrier impact crater; Determining the remote sensing image to which the target barrier impact crater belongs based on the position information includes: Converting the longitude and latitude coordinates of the target barrier crater to the coordinates of the Martian sphere where the target barrier crater is located to obtain the converted coordinates of the target barrier crater; Based on the converted coordinates, the map range of the target barrier crater on the Martian sphere is determined, and a remote sensing image matching the map range is searched in a preset remote sensing image database, wherein the preset remote sensing image database stores remote sensing images corresponding to different map ranges.

[0007] Optionally, matching a digital elevation model homologous to the remote sensing image includes: Based on the position information of the target barrier crater, determining an initial digital elevation model corresponding to the target barrier crater; Projecting the remote sensing image and the initial digital elevation model onto the same coordinate system, and processing the projected remote sensing image and the projected initial digital elevation model to have the same resolution, thereby obtaining the processed remote sensing image and the processed initial digital elevation model; The feature points of the target barrier crater are matched between the processed remote sensing image and the processed initial digital elevation model, and based on the matching result, a digital elevation model homologous to the remote sensing image is identified in the initial digital elevation model.

[0008] Optionally, segmenting the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model includes: Obtaining prompt information for segmenting the remote sensing image; Acquire a remote sensing large model, wherein the remote sensing large model includes a multi-head self-attention mechanism for extracting impact crater features, a feedforward neural network for transforming impact crater features, an encoder for encoding prompt information, and an image segmentation network for segmenting remote sensing images; The remote sensing image, the digital elevation model, and the prompt information are inputted into the remote sensing large model together, the crater features are extracted from the remote sensing image and the digital elevation model by the multi-head self-attention mechanism to obtain crater features, the crater features are nonlinearly transformed by the feedforward neural network to obtain nonlinear crater features, the prompt information is encoded by the encoder to obtain encoding features, the nonlinear crater features and the encoding features are image segmented by the image segmentation network to obtain the contour area of ​​the target barrier crater and its corresponding sputtering object; After segmenting the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model, the method further includes: If the contour area is a label mask file, the label mask file is converted into a vector file containing the geometric feature parameters of the contour area using a preset conversion method, wherein the method of converting the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method includes: Converting the label mask file into a binary image, and processing the binary image into a raster image; Determining a geographic transformation matrix of the raster image, and extracting an affine matrix for transforming from pixel coordinates to geographic coordinates from the geographic transformation matrix; The inverse matrix of the affine matrix is ​​calculated, and based on the inverse matrix, the Martian geographic coordinates of the contour area are converted into pixel coordinates, and based on the contour area after the pixel coordinates are converted, a vector file containing the geometric feature parameters of the contour area is determined.

[0009] Optionally, the geometric characteristic parameters include the pit area and pit radius of the target barrier impact pit in the contour area, and the sputtering perimeter and sputtering area of ​​the sputtering seat formed by each of the sputtering objects; The step of evaluating the morphological characteristics of the target barrier crater based on the geometric characteristic parameters includes: Determining a flow coefficient of the sputtered material corresponding to the target barrier impact crater based on the crater area, the sputtering area, and the crater radius; Determining a lobation factor of the target barrier impact crater based on the sputtering perimeter and the sputtering area; Based on the flow coefficient and the lobation coefficient, the morphological characteristics of the target barrier impact crater are determined.

[0010] Optionally, before matching the digital elevation model homologous to the remote sensing image, the method further includes: Filtering the remote sensing image using a preset filter, and projecting the filtered remote sensing image to the Martian geographic coordinate system where the target barrier impact crater is located, to obtain the projected remote sensing image; Determining the center position and the extension range of the target barrier crater, and based on the center position and the extension range, cutting out a target remote sensing image corresponding to the target barrier crater from the projected remote sensing image; The matching of the digital elevation model homologous to the remote sensing image comprises: Matching a digital elevation model that is homologous to the target remote sensing image.

[0011] Optionally, after segmenting the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model, the method further comprises: Taking the contour area as the contour area to be verified, obtaining the reference contour area of ​​the target barrier impact crater and its corresponding sputtering product segmented in the remote sensing image, and determining the reference contour area and reference contour perimeter of the reference contour area, and the contour area to be verified and the contour perimeter to be verified of the contour area to be verified; Calculating the area overlap between the reference contour area and the contour area to be verified, and determining the perimeter detection parameter based on the reference contour perimeter and the contour perimeter to be verified, and determining the area detection parameter based on the reference contour area and the contour area to be verified; Based on the region overlap, the perimeter detection parameters, and the area detection parameters, determine whether the contour area to be verified meets the segmentation requirements. If so, determine the morphological features of the target barrier impact crater based on the contour area to be verified. Otherwise, re-segment the contour area of ​​the target barrier impact crater and its corresponding spatter in the remote sensing image.

[0012] According to a second aspect of the present invention, there is provided a device for evaluating the morphological characteristics of a Martian rampart impact crater, comprising: an acquisition unit, configured to acquire position information of the target barrier impact crater in response to a morphological feature evaluation signal of the target barrier impact crater; A matching unit, configured to determine the remote sensing image to which the target barrier crater belongs based on the position information, and to match a digital elevation model homologous to the remote sensing image; A segmentation unit, configured to segment the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model; An evaluation unit is used to determine geometric characteristic parameters of the contour area and evaluate the morphological characteristics of the target barrier crater based on the geometric characteristic parameters.

[0013] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for evaluating the morphological characteristics of Martian rampart impact craters.

[0014] According to a fourth aspect of the present invention, there is provided a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for evaluating the morphological characteristics of Martian rampart impact craters when executing the program.

[0015] According to a method, device and storage medium for evaluating the morphological characteristics of a Martian rampart impact crater provided by the present invention, compared with the current method of identifying the morphological characteristics of a Martian rampart impact crater by human naked eyes, the present invention obtains the position information of the target rampart impact crater by responding to the morphological characteristic evaluation signal of the target rampart impact crater; and based on the position information, determines the remote sensing image to which the target rampart impact crater belongs, and matches the digital elevation model homologous to the remote sensing image; then, based on the remote sensing image and the digital elevation model, segments the contour area of ​​the target rampart impact crater and its corresponding sputtering in the remote sensing image; finally determines the geometric characteristic parameters of the contour area, and evaluates the morphological characteristics of the target rampart impact crater based on the geometric characteristic parameters. Thus, through the remote sensing image and the digital elevation model to which the target rampart impact crater belongs, the contour area of ​​the target rampart impact crater and its corresponding sputtering is segmented in the remote sensing image, and finally, according to the geometric characteristic parameters of the contour area, the morphological characteristics of the target rampart impact crater are evaluated, which can realize the automatic evaluation of the morphological characteristics of the rampart impact crater, avoid manual participation, and thus improve the evaluation efficiency and evaluation accuracy of the morphological characteristics of the Martian rampart impact crater. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 A flow chart of a method for evaluating the morphological characteristics of a Martian rampart impact crater provided by an embodiment of the present invention is shown; Figure 2 A flow chart of another method for evaluating the morphological characteristics of a Martian rampart impact crater provided by an embodiment of the present invention is shown; Figure 3 A schematic structural diagram of a device for evaluating the morphological characteristics of a Martian rampart impact crater provided by an embodiment of the present invention is shown; Figure 4 A schematic structural diagram of another device for evaluating the morphological characteristics of a Martian rampart impact crater provided by an embodiment of the present invention is shown; Figure 5 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0018] At present, the method of identifying the morphological characteristics of the Martian Rampart crater with the naked eye is greatly affected by human subjective factors, which will result in large errors in the assessment of the morphological characteristics of the Martian Rampart crater. At the same time, manual assessment is time-consuming and labor-intensive.

[0019] In order to solve the above problems, an embodiment of the present invention provides a method for evaluating the morphological characteristics of a Martian rampart impact crater, such as Figure 1 As shown, the method includes: 101. In response to a morphological feature evaluation signal of a target barrier impact crater, obtain position information of the target barrier impact crater.

[0020] The target barrier crater may be any barrier crater distributed on the surface of Mars; the location information is the latitude and longitude coordinates of the target barrier crater. The number of target barrier craters may be one or more, and the embodiment of the present invention does not specifically limit the number of target barrier craters.

[0021] For the embodiment of the present invention, when the morphological characteristic evaluation signal of the target barrier crater is received, the longitude and latitude coordinates of the target barrier crater are determined through geological survey data, geographic information system and other software.

[0022] 102. Based on the location information, determine the remote sensing image to which the target barrier crater belongs, and match the digital elevation model that is homologous to the remote sensing image.

[0023] For the embodiment of the present invention, in order to evaluate the morphological characteristics of the target barrier crater, it is first necessary to determine the remote sensing image to which the target barrier crater belongs. Based on this, step 102 specifically includes: converting the latitude and longitude coordinates of the target barrier crater to the coordinates of the Martian sphere where the target barrier crater is located, and obtaining the converted coordinates of the target barrier crater; based on the converted coordinates, determining the map range of the target barrier crater on the Martian sphere, and searching for remote sensing images that match the map range in a preset remote sensing image database, wherein the preset remote sensing image database stores remote sensing images corresponding to different map ranges.

[0024] Among them, the Mars spherical coordinates are the coordinates on the Mars coordinate system. The Mars coordinate system is a coordinate system based on the shape, size and rotation characteristics of Mars. Specifically, the longitude coordinate conversion is performed according to the following formula:

[0025] in, is the converted coordinate corresponding to the longitude of the target rampart crater, is the longitude coordinate of the target Rampart crater, is the radius of the Mars sphere where the target rampart impact crater is located. Further, the latitude coordinate conversion is performed according to the following formula:

[0026] in, is the converted coordinate corresponding to the latitude of the target rampart crater, is the latitude coordinate of the target Rampart crater, is the radius of the Mars sphere where the target barrier crater is located. The converted coordinates of the target barrier crater can be obtained through the above two formulas. Then, according to the converted coordinates, the map range of the target barrier crater on the Mars sphere is determined, and the search range is determined based on the map range. Finally, the remote sensing image to which the target barrier crater belongs is searched in the preset remote sensing image database according to the search range. For example, the converted longitude coordinate of the target barrier crater is 153.4°, and the converted latitude coordinate is 20.5°. The longitude range searched is 152° to 154°, and the latitude range searched is -20° to -22°. Write an automated search script, execute the script and finally output the remote sensing image within the above longitude and latitude range, and determine the remote sensing image as the remote sensing image to which the target barrier crater belongs. Then, the searched remote sensing image is verified to verify the search accuracy of the remote sensing image.

[0027] Furthermore, after determining the remote sensing image to which the target barrier crater belongs, it is also necessary to match a digital elevation model homologous to the remote sensing image. Based on this, the method includes: determining an initial digital elevation model corresponding to the target barrier crater based on the position information of the target barrier crater; projecting the remote sensing image and the initial digital elevation model onto the same coordinate system, and processing the projected remote sensing image and the projected initial digital elevation model to the same resolution to obtain the processed remote sensing image and the processed initial digital elevation model; matching feature points of the target barrier crater in the processed remote sensing image and the processed initial digital elevation model, and identifying a digital elevation model homologous to the remote sensing image in the initial digital elevation model based on the matching result.

[0028] Among them, the characteristic points of the target barrier impact crater may be edge points, corner points and other points with significant characteristics of the target barrier impact crater.

[0029] Specifically, first ensure that the coordinate system of the target barrier crater is consistent with the coordinate system of the digital elevation model data source, then determine the initial digital elevation model that can cover the target barrier crater in the digital elevation model data source according to the latitude and longitude coordinates of the target barrier crater (wherein, the digital elevation model data source stores multiple digital elevation model data), then convert the initial digital elevation model and the remote sensing image into the same projection coordinate system, and process the projected remote sensing image and the projected initial digital elevation model to the same resolution, for example, process the resolution of the remote sensing image to the same resolution image as the initial digital elevation model. Then use the feature extraction algorithm to extract the feature points of the target barrier crater in the processed remote sensing image and the processed initial digital elevation model respectively, and match the feature points extracted from the remote sensing image with the feature points extracted from the initial digital elevation model to find matching feature points in the initial digital elevation model that match the feature points in the remote sensing image, and finally determine the area corresponding to each matching feature point in the initial digital elevation model, and determine the digital elevation model under the area as the digital elevation model of the same source as the remote sensing image. Finally, the morphological characteristics of the barrier crater are estimated based on remote sensing images and digital elevation models, which can realize the automation of the estimation of the morphological characteristics of the barrier crater and avoid the time consumed by manual participation and estimation errors, thereby improving the estimation accuracy and efficiency of the morphological characteristics of the barrier crater.

[0030] 103. Based on remote sensing images and digital elevation models, the target barrier impact crater and its corresponding splash area are segmented in the remote sensing image.

[0031] The spatter of the barrier crater refers to the material excavated and ejected from the periphery of the barrier crater due to the impact of the high-speed impact body during the formation of the barrier crater. The contour area in the embodiment of the present invention is the area corresponding to the target barrier crater containing the spatter.

[0032] For the embodiment of the present invention, the remote sensing image is first geometrically corrected and radiometrically corrected to ensure the accuracy and readability of the image, to ensure that the resolution of the digital elevation model matches the remote sensing image, or to perform appropriate resampling to match the resolution. The remote sensing image and the digital elevation model are then used as multi-source inputs, and image fusion or data assimilation techniques are used to integrate the information of the two. An image segmentation algorithm is applied to more accurately identify the contour areas of the impact craters and ejecta. The segmented contour areas are then smoothed to reduce the effects of noise and irregular edges. Morphological operations (such as dilation and erosion) can be used to further adjust the shape and size of the contour area, thereby ensuring the extraction accuracy of the contour area and further improving the estimation accuracy of the morphological features of the barrier impact craters.

[0033] 104. Determine the geometric characteristic parameters of the contour area, and based on the geometric characteristic parameters, evaluate the morphological characteristics of the target barrier crater.

[0034] For an embodiment of the present invention, after extracting the contour area of ​​the target barrier impact crater and its corresponding sputtering product, if the contour area is a label mask file, it is also necessary to perform file format conversion on the label mask file. Based on this, the method includes: if the contour area is a label mask file, converting the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method, wherein the method of converting the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method includes: converting the label mask file into a binary image, and processing the binary image into a raster image; determining the geographic transformation matrix of the raster image, and extracting an affine matrix for converting from pixel coordinates to geographic coordinates from the geographic transformation matrix; calculating the inverse matrix of the affine matrix, and based on the inverse matrix, converting the Martian geographic coordinates of the contour area into pixel coordinates, and determining the vector file containing the geometric feature parameters of the contour area based on the contour area after the pixel coordinates are converted.

[0035] The label in the label mask file is the unique identifier of the contour area, the mask file is an image file containing mask information, and the mask refers to the "mask" covered on the layer. It allows you to create complex image effects by controlling the visibility of the layer without directly modifying the layer content. The vector file refers to the contour area file described in digital form. The preset conversion method is the method selected according to actual needs, such as the conversion method of the geographic conversion matrix.

[0036] In the embodiment of the present invention, the contour area image under the label mask file is binarized to obtain a binary image, and then the binary image is processed into a raster image, and then the geographic transformation matrix of the raster image is determined. The geographic transformation matrix is ​​determined by: according to the geographic coordinates of the upper left corner pixel in the raster image , the horizontal resolution of the upper left pixel , the horizontal rotation component of the raster image , the geographic coordinates of the upper right pixel of the raster image , the vertical resolution of the upper right corner pixel , the vertical rotation component of the raster image , determine the geographic transformation matrix of the raster image As shown below:

[0037] Furthermore, the affine matrix A for transforming from pixel coordinates to geographic coordinates is extracted from the geographic transformation matrix as follows:

[0038] On this basis, determine the inverse matrix of the affine matrix A As shown below:

[0039] in, Furthermore, according to the above inverse matrix, the Mars geographic coordinates of the contour area are converted to pixel coordinates. The conversion formula is as follows:

[0040] in, is the pixel coordinate of the contour area, are the Martian geographic coordinates of the contour area, are the geographic coordinates of the upper left pixel in the raster image, is the geographic coordinate of the pixel in the upper right corner of the raster image. In this way, the contour area after the coordinate conversion can be obtained, and the geometric feature parameters of the contour area can be determined in the remote sensing image and the digital elevation model, and finally the contour area containing the geometric feature parameters can be constructed as a vector file. If the morphological characteristics of the barrier crater are to be estimated later, the geometric feature parameters of the contour area corresponding to the barrier crater can be directly obtained in the vector file, thereby improving the estimation efficiency of the morphological characteristics of the barrier crater. Therefore, through the remote sensing image and digital elevation model to which the target barrier crater belongs, the contour area of ​​the target barrier crater and its corresponding sputtering are segmented in the remote sensing image, and finally the morphological characteristics of the target barrier crater are evaluated according to the geometric feature parameters of the contour area, which can realize the automatic evaluation of the morphological characteristics of the barrier crater and avoid manual participation, thereby improving the evaluation efficiency and accuracy of the morphological characteristics of the Martian barrier crater.

[0041] According to a method for evaluating the morphological characteristics of a Martian rampart crater provided by the present invention, compared with the current method of identifying the morphological characteristics of a Martian rampart crater by human naked eyes, the present invention obtains the position information of the target rampart crater by responding to the morphological characteristic evaluation signal of the target rampart crater; and based on the position information, determines the remote sensing image to which the target rampart crater belongs, and matches the digital elevation model homologous to the remote sensing image; then, based on the remote sensing image and the digital elevation model, segments the contour area of ​​the target rampart crater and its corresponding sputtering in the remote sensing image; finally determines the geometric characteristic parameters of the contour area, and evaluates the morphological characteristics of the target rampart crater based on the geometric characteristic parameters. Thus, through the remote sensing image and the digital elevation model to which the target rampart crater belongs, the contour area of ​​the target rampart crater and its corresponding sputtering is segmented in the remote sensing image, and finally, according to the geometric characteristic parameters of the contour area, the morphological characteristics of the target rampart crater are evaluated, which can realize the automatic evaluation of the morphological characteristics of the rampart crater and avoid manual participation, thereby improving the evaluation efficiency and evaluation accuracy of the morphological characteristics of the Martian rampart crater.

[0042] Further, in order to better illustrate the above process of evaluating the morphological characteristics of the Martian rampart impact crater, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another method for evaluating the morphological characteristics of the Martian rampart impact crater, such as Figure 2 As shown, the method includes: 201. In response to a morphological feature evaluation signal of a target barrier impact crater, obtain position information of the target barrier impact crater.

[0043] Specifically, when a morphological feature evaluation signal of a target barrier crater is received, the latitude and longitude coordinate information of the target barrier crater is acquired.

[0044] 202. Based on the location information, determine the remote sensing image to which the target barrier crater belongs, and match the digital elevation model that is homologous to the remote sensing image.

[0045] Specifically, according to the latitude and longitude coordinates of the target barrier crater, a remote sensing image that can cover the target barrier crater is searched in a preset remote sensing image database, and then the remote sensing image is preprocessed. Based on this, the method includes: filtering the remote sensing image using a preset filter, and projecting the filtered remote sensing image to the Martian geographic coordinate system where the target barrier crater is located to obtain the projected remote sensing image; determining the center position and extension range of the target barrier crater, and based on the center position and the extension range, cutting out the target remote sensing image corresponding to the target barrier crater in the projected remote sensing image. Finally, matching the digital elevation model homologous to the target remote sensing image.

[0046] Among them, the preset filter can be at least one of a low-pass filter, a high-pass filter, a band-pass filter, etc. Specifically, before filtering, the remote sensing image is preprocessed, including removing redundant information, correcting geometric distortion, etc., and the preprocessed remote sensing image is filtered using the preset filter, and then the remote sensing image coordinate systems of different data sources are unified into the same coordinate system, and the equiangular projection is adopted, and the Mars radius datum is used as the unified datum, and the remote sensing image is projected onto the Mars geographic coordinate system to obtain the projected remote sensing image. For example, the gdalwarp (GDAL command line tool) tool of GDAL (Geospatial Data Abstraction Library, a geospatial data conversion library) can be used for image reprojection and the pyproj (library for coordinate projection and conversion of geospatial data) library of Python (a programming language for coordinate conversion) can be used to complete the coordinate conversion. The cropping boundary is calculated using the center coordinates and extended range of the barrier crater, with reference to the center position of the barrier crater in the barrier crater database file, and the outermost visual boundary of the barrier crater is used as the starting boundary. For example, the length and width directions are expanded outward by about 3 times the maximum distance of the splash, and the barrier crater is ensured to be located in the center of the image, so as to crop out the target remote sensing image corresponding to the target barrier crater in the projected remote sensing image. This method is conducive to reducing image deformation, especially in high latitudes and image edge areas, and can effectively reduce the residual caused by projection. For example, the long side range of the cropping boundary L=2×(3×R), where L is the long side range and R is the maximum splash distance of the splash of the target barrier crater. The wide side range of the cropping boundary W=2×(3×R), where W is the wide side range and R is the maximum splash distance of the splash of the target barrier crater. The coordinates of the upper left corner of the target remote sensing image are , the coordinates of the lower right corner are ,in, is the horizontal coordinate of the center of the target barrier crater, is the ordinate of the center of the target barrier crater, is the long side range of the clipping boundary, is the wide range of the cropping boundary. On this basis, the effective cropped remote sensing image of the target rampart crater is obtained, that is, the target remote sensing image corresponding to the target rampart crater. Then, according to the latitude and longitude coordinates of the target rampart crater, the initial digital elevation model is obtained, and the overlapping area with the target remote sensing image is determined in the initial digital elevation model, and the elevation model corresponding to the overlapping area is determined as the digital elevation model of the same source as the target remote sensing image.

[0047] 203. Obtain prompt information for segmenting the remote sensing image.

[0048] Among them, the remote sensing image is the target remote sensing image referred to in step 202; the prompt information refers to the guiding information for accurately segmenting the remote sensing image. For example, the prompt information can be information such as the user clicking one or more points, annotation boxes, texts, etc. on the remote sensing image to mark the representative pixel positions of the spatter area. In the segmentation process, the guidance starts from the marked points and extracts the spatter area related to these points. The one or more points can be the points at the center of the impact crater and the edge of the spatter, so as to prompt the distinction between the impact crater body and the spatter during the image segmentation process, so that by introducing the prompt information, the image segmentation accuracy can be improved, thereby improving the estimation accuracy of the morphological characteristics of the barrier impact crater.

[0049] 204. Obtain a large remote sensing model, wherein the large remote sensing model includes a multi-head self-attention mechanism for extracting crater features, a feed-forward neural network for transforming crater features, an encoder for encoding prompt information, and an image segmentation network for segmenting remote sensing images.

[0050] 205. The remote sensing image, digital elevation model and prompt information are input into the remote sensing large model together. The impact crater features are extracted from the remote sensing image and the digital elevation model through the multi-head self-attention mechanism to obtain the impact crater features. The impact crater features are nonlinearly transformed through a feedforward neural network to obtain nonlinear impact crater features. The prompt information is encoded through an encoder to obtain encoded features. The nonlinear impact crater features and the encoded features are image segmented through an image segmentation network to obtain the contour area of ​​the target barrier impact crater and its corresponding spatter.

[0051] Specifically, the remote sensing large model is composed of a multi-head self-attention mechanism, a feedforward neural network, an encoder, an image segmentation network, etc. In order to improve the segmentation accuracy of the remote sensing large model, it is first necessary to train and test the initial remote sensing large model using a sample data set consisting of remote sensing images and digital elevation models with segmentation labels, so as to obtain a remote sensing large model with segmentation accuracy that meets the requirements. Furthermore, the remote sensing images and digital elevation models are processed by image enhancement, resolution optimization, and standardization operations to enhance the large model's ability to analyze remote sensing images and digital elevation models and visual effects. After that, the processed remote sensing images, digital elevation models, and prompt information are input into the remote sensing large model together, and the remote sensing images and digital elevation models are feature extracted through the multi-head self-attention mechanism in the remote sensing large model to obtain the image features corresponding to the remote sensing images and the elevation features corresponding to the digital elevation models. The feature extraction formula of the multi-head self-attention mechanism is as follows:

[0052] Among them, Q, K, and V represent the query vector, key vector, and value of the remote sensing image and digital elevation model, respectively. is the dimension of the key, T represents the transpose of K, is the remote sensing feature corresponding to the remote sensing image, or the elevation feature corresponding to the digital elevation model. is the feature extraction function. Through the above formula, the remote sensing large model can focus on different parts of the remote sensing image and the digital elevation model, thereby achieving more accurate feature extraction.

[0053] Furthermore, the remote sensing features and elevation features are fused to obtain the crater features, which are then input into the feedforward neural network for feature transformation to obtain nonlinear crater features. The transformation formula for feature transformation is as follows:

[0054] in, is the activation function of the feedforward neural network layer, and are the weight matrices of the feedforward neural network layer, and are the bias terms of the feedforward neural network layer, The characteristics of the impact crater, The nonlinear crater feature is used to transform the crater feature nonlinearly through the feedforward neural network, which can enhance the expression ability of the remote sensing large model and thus improve the image segmentation accuracy.

[0055] At the same time, the prompt information is encoded by the encoder in the remote sensing large model to obtain the encoding features, that is, the encoder converts the information such as points, boxes or text in the prompt information into feature display. The encoding features and nonlinear impact crater features are then fused to obtain fused features, and finally the fused features are input into the image segmentation network, and the target barrier impact crater and its corresponding sputtering object contour area are segmented through the image segmentation network. The embodiment of the present invention introduces prompt information in the image segmentation process, so that the high-level image features can obtain the directional guidance of the user input, thereby realizing the accurate segmentation of the barrier impact crater and its corresponding sputtering object contour.

[0056] Furthermore, after the contour area of ​​the target barrier impact crater and its corresponding sputtering product is segmented in the remote sensing image, in order to improve the estimation accuracy of the morphological characteristics of the target barrier impact crater, it is also necessary to verify whether the segmentation of the contour area is accurate or not. Based on this, the method includes: taking the contour area as the contour area to be verified, obtaining the reference contour area of ​​the target barrier impact crater and its corresponding sputtering product segmented in the remote sensing image, and determining the reference contour area and reference contour perimeter of the reference contour area, the contour area to be verified and the contour perimeter to be verified of the contour area to be verified; calculating the reference contour area; The area overlap between the region and the contour area to be verified, and based on the reference contour perimeter and the contour perimeter to be verified, the perimeter detection parameters are determined; based on the reference contour area and the contour area to be verified, the area detection parameters are determined; based on the area overlap, the perimeter detection parameters, and the area detection parameters, whether the contour area to be verified meets the segmentation requirements is judged; if so, the morphological features of the target barrier impact crater are determined based on the contour area to be verified; otherwise, the contour area of ​​the target barrier impact crater and its corresponding sputtering object are segmented again in the remote sensing image.

[0057] Specifically, the contour area is used as the contour area to be verified, and the reference contour area of ​​the target barrier impact crater and its corresponding sputtering product is obtained at the same time. The reference contour area can be manually drawn by an expert. Then the reference contour area of ​​the reference contour area is determined. and the reference contour perimeter , the contour area to be verified and the perimeter of the contour to be verified , and calculate the regional overlap according to the following formula :

[0058] in, is the overlapping area between the reference contour area and the contour area to be verified, is the area of ​​the union of the reference contour area and the contour area to be verified. At the same time, the perimeter detection parameter is calculated according to the following formula: :

[0059] At the same time, the area detection parameters are calculated according to the following formula: :

[0060] Further, it is determined whether the regional overlap is less than the preset overlap threshold, whether the perimeter detection parameter is less than the preset perimeter threshold, and whether the area detection parameter is less than the preset area threshold; if the regional overlap is less than the preset overlap threshold, and / or the perimeter detection parameter is less than the preset perimeter threshold, and / or the area detection parameter is less than the preset area threshold, then it is determined that the contour area extraction of the target barrier impact crater and its corresponding sputtering object does not meet the requirements, and at this time it is necessary to re-extract the contour area of ​​the target barrier impact crater and its corresponding sputtering object in the remote sensing image; if the regional overlap is greater than or equal to the preset overlap threshold, and the perimeter detection parameter is greater than or equal to the preset perimeter threshold, and the area detection parameter is greater than or equal to the preset area threshold, then it is determined that the contour area extraction of the target barrier impact crater and its corresponding sputtering object meets the requirements, and there is no need to re-extract the contour area, wherein the preset overlap threshold, the preset perimeter threshold, and the preset area threshold are all values ​​set according to actual needs. The embodiment of the present invention can ensure the extraction accuracy of the contour area by judging whether the contour area extraction meets the requirements, thereby improving the evaluation accuracy of the morphological characteristics of the barrier impact crater.

[0061] 206. Determine the geometric characteristic parameters of the contour area, and based on the geometric characteristic parameters, evaluate the morphological characteristics of the target barrier crater.

[0062] The geometric characteristic parameters include the pit area and pit radius of the target barrier impact crater in the contour area, and the sputtering perimeter and sputtering area of ​​the sputtering seat formed by each sputtering object. For the embodiment of the present invention, the geometric characteristic parameters of the contour area can be extracted in the remote sensing image and the digital elevation model, and then the morphological characteristics of the target barrier impact crater need to be evaluated based on the geometric characteristic parameters. Based on this, step 206 specifically includes: determining the flow coefficient of the sputtering object corresponding to the target barrier impact crater based on the pit area, the sputtering area, and the pit radius; determining the lobate coefficient of the target barrier impact crater based on the sputtering perimeter and the sputtering area; determining the morphological characteristics of the target barrier impact crater based on the flow coefficient and the lobate coefficient.

[0063] Specifically, the flow coefficient of the sputtering material corresponding to the target barrier impact crater is calculated according to the following formula: :

[0064] in, is the sputtering area of ​​the sputtering seat (the total area of ​​each sputtered object), is the crater area of ​​the target barrier impact crater, is the radius of the target barrier crater. The flow distance of continuous ejecta in the flow coefficient may be affected by the slope / roughness of the surface before impact, the viscosity of the ejecta flow, the size distribution of debris in the ejecta, and the impact crater velocity, which may lead to uneven diffusion of continuous ejecta. In order to ensure the scientificity and reliability of the flow coefficient calculation, the barrier crater and its ejecta area are calculated by directly extracting the contour, which reduces the error caused by the error in the average distance calculation. The batch calculation process can be implemented in a programming language, and the processing efficiency can be improved through multi-threaded calculation, and visual results can be generated, including flow area distribution diagrams and parameter statistics tables.

[0065] At the same time, the lobate coefficient of the target barrier crater is calculated according to the following formula: :

[0066] in, is the sputtering perimeter of the sputtering seat, is the sputtering area of ​​the sputtering mat. Finally, the morphological characteristics of the target barrier impact crater are determined based on the flow coefficient and the lobate coefficient. For example, the larger the flow coefficient, the stronger the fluidity of the sputtering material and the wider the distribution range. Conversely, it indicates that the flow of the sputtering material is restricted. When the lobate coefficient is closer to 1, the continuous sputtering material is more regular and close to a circle. The larger the lobate coefficient, the more irregular the appearance of the continuous sputtering material.

[0067] Furthermore, if the flow coefficients and lobation coefficients of multiple target barrier craters are determined, the change curves of the flow coefficients and lobation coefficients of all target barrier craters can be plotted, and statistical reports can be generated to analyze the regular characteristics of the flow coefficients and lobation coefficients. This can reveal the radial and circumferential expansion and distribution of the spatters of barrier craters, and provide basic data for subsequent simulations of impact events.

[0068] According to another method for evaluating the morphological characteristics of a Martian rampart impact crater provided by the present invention, compared with the current method of identifying the morphological characteristics of a Martian rampart impact crater by human naked eyes, the present invention obtains the position information of the target rampart impact crater by responding to the morphological characteristic evaluation signal of the target rampart impact crater; and based on the position information, determines the remote sensing image to which the target rampart impact crater belongs, and matches the digital elevation model homologous to the remote sensing image; then, based on the remote sensing image and the digital elevation model, segments the contour area of ​​the target rampart impact crater and its corresponding sputtering in the remote sensing image; finally determines the geometric characteristic parameters of the contour area, and evaluates the morphological characteristics of the target rampart impact crater based on the geometric characteristic parameters. Thus, through the remote sensing image and the digital elevation model to which the target rampart impact crater belongs, the contour area of ​​the target rampart impact crater and its corresponding sputtering is segmented in the remote sensing image, and finally, according to the geometric characteristic parameters of the contour area, the morphological characteristics of the target rampart impact crater are evaluated, which can realize the automatic evaluation of the morphological characteristics of the rampart impact crater, avoid manual participation, and thus improve the evaluation efficiency and evaluation accuracy of the morphological characteristics of the Martian rampart impact crater.

[0069] Further, as Figure 1 In a specific implementation, an embodiment of the present invention provides a device for evaluating the morphological characteristics of a Martian rampart impact crater, such as Figure 3 As shown, the device includes: an acquisition unit 31, a matching unit 32, a segmentation unit 33, and an evaluation unit 34.

[0070] The acquisition unit 31 may be used to acquire the position information of the target barrier impact crater in response to the morphological feature evaluation signal of the target barrier impact crater.

[0071] The matching unit 32 may be used to determine the remote sensing image to which the target barrier crater belongs based on the position information, and to match a digital elevation model homologous to the remote sensing image.

[0072] The segmentation unit 33 may be used to segment the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model.

[0073] The evaluation unit 34 may be used to determine geometric characteristic parameters of the contour area, and based on the geometric characteristic parameters, evaluate the morphological characteristics of the target barrier crater.

[0074] In a specific application scenario, the location information is the latitude and longitude coordinates of the target barrier crater. In order to determine the remote sensing image to which the target barrier crater belongs, such as Figure 4 As shown, the matching unit 32 includes a coordinate conversion module 321 and a search module 322 .

[0075] The coordinate conversion module 321 can be used to convert the longitude and latitude coordinates of the target barrier crater into the coordinates of the Martian sphere where the target barrier crater is located, so as to obtain the converted coordinates of the target barrier crater.

[0076] The search module 322 can be used to determine the map range of the target barrier crater on the Martian sphere based on the converted coordinates, and search for remote sensing images matching the map range in a preset remote sensing image database, wherein the preset remote sensing image database stores remote sensing images corresponding to different map ranges.

[0077] In a specific application scenario, in order to match a digital elevation model that is homologous to a remote sensing image, the matching unit 32 further includes a first determination module 323 , a processing module 324 , and a matching module 325 .

[0078] The first determination module 323 may be used to determine an initial digital elevation model corresponding to the target barrier crater based on the position information of the target barrier crater.

[0079] The processing module 324 can be used to project the remote sensing image and the initial digital elevation model onto the same coordinate system, and process the projected remote sensing image and the projected initial digital elevation model to the same resolution to obtain the processed remote sensing image and the processed initial digital elevation model.

[0080] The matching module 325 can be used to match the feature points of the target barrier crater in the processed remote sensing image and the processed initial digital elevation model, and based on the matching result, identify the digital elevation model homologous to the remote sensing image in the initial digital elevation model.

[0081] In a specific application scenario, in order to segment the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image, the segmentation unit 33 includes an acquisition module 331 and a segmentation module 332 .

[0082] The acquisition module 331 may be used to acquire prompt information for segmenting the remote sensing image.

[0083] The acquisition module 331 can also be used to acquire a large remote sensing model, wherein the large remote sensing model includes a multi-head self-attention mechanism for extracting crater features, a feedforward neural network for transforming crater features, an encoder for encoding prompt information, and an image segmentation network for segmenting remote sensing images.

[0084] The segmentation module 332 can be used to input the remote sensing image, the digital elevation model, and the prompt information into the remote sensing large model, extract crater features from the remote sensing image and the digital elevation model through the multi-head self-attention mechanism to obtain crater features, perform nonlinear transformation on the crater features through the feedforward neural network to obtain nonlinear crater features, encode the prompt information through the encoder to obtain encoded features, perform image segmentation on the nonlinear crater features and the encoded features through the image segmentation network to obtain the contour area of ​​the target barrier crater and its corresponding spatter.

[0085] In a specific application scenario, in order to convert the file format of the contour area, the device further includes a file conversion unit 35 .

[0086] The file conversion unit 35 can be used to convert the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method if the contour area is a label mask file, wherein the method of converting the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method includes: converting the label mask file into a binary image, and processing the binary image into a raster image; determining the geographic transformation matrix of the raster image, and extracting an affine matrix for converting from pixel coordinates to geographic coordinates from the geographic transformation matrix; calculating the inverse matrix of the affine matrix, and based on the inverse matrix, converting the Martian geographic coordinates of the contour area into pixel coordinates, and determining the vector file containing the geometric feature parameters of the contour area based on the contour area after the pixel coordinates are converted.

[0087] In a specific application scenario, the geometric feature parameters include the pit area and pit radius of the target barrier impact crater in the contour area, and the sputtering circumference and sputtering area of ​​the sputtering seat formed by each of the sputtering objects; in order to evaluate the morphological characteristics of the target barrier impact crater, the evaluation unit 34 can be specifically used to determine the flow coefficient of the sputtering object corresponding to the target barrier impact crater based on the pit area, the sputtering area, and the pit radius; determine the lobate coefficient of the target barrier impact crater based on the sputtering circumference and the sputtering area; and determine the morphological characteristics of the target barrier impact crater based on the flow coefficient and the lobate coefficient.

[0088] In a specific application scenario, in order to pre-process the remote sensing image, the device further includes a pre-processing unit 36 ​​.

[0089] The preprocessing unit 36 ​​can be specifically used to filter the remote sensing image using a preset filter, and project the filtered remote sensing image to the Martian geographic coordinate system where the target barrier crater is located to obtain the projected remote sensing image; determine the center position and expansion range of the target barrier crater, and based on the center position and the expansion range, crop the target remote sensing image corresponding to the target barrier crater in the projected remote sensing image.

[0090] The matching unit 32 may also be used to match a digital elevation model that is homologous to the target remote sensing image.

[0091] In a specific application scenario, in order to verify the segmentation accuracy of the contour area, the device further includes a segmentation verification unit 37 .

[0092] The segmentation verification unit 37 can be used to take the contour area as the contour area to be verified, obtain the reference contour area of ​​the target barrier impact crater and its corresponding sputtering product segmented in the remote sensing image, and determine the reference contour area and reference contour perimeter of the reference contour area, the contour area to be verified and the contour perimeter to be verified of the contour area to be verified; calculate the area overlap between the reference contour area and the contour area to be verified, and determine the perimeter detection parameters based on the reference contour perimeter and the contour perimeter to be verified, and determine the area detection parameters based on the reference contour area and the contour area to be verified; based on the area overlap, the perimeter detection parameters, and the area detection parameters, determine whether the contour area to be verified meets the segmentation requirements; if so, determine the morphological characteristics of the target barrier impact crater based on the contour area to be verified; otherwise, re-segment the contour area of ​​the target barrier impact crater and its corresponding sputtering product in the remote sensing image.

[0093] It should be noted that for other corresponding descriptions of the functional modules involved in the device for evaluating the morphological characteristics of a Martian rampart impact crater provided in an embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0094] Based on the above Figure 1The method shown, accordingly, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are implemented: in response to a morphological feature evaluation signal of a target barrier crater, obtaining the position information of the target barrier crater; based on the position information, determining the remote sensing image to which the target barrier crater belongs, and matching a digital elevation model homologous to the remote sensing image; based on the remote sensing image and the digital elevation model, segmenting the contour area of ​​the target barrier crater and its corresponding sputtering in the remote sensing image; determining the geometric feature parameters of the contour area, and based on the geometric feature parameters, evaluating the morphological features of the target barrier crater.

[0095] Based on the above Figure 1 The method shown and Figure 3 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 5 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, and when the processor 41 executes the program, the following steps are implemented: in response to a morphological feature evaluation signal of a target barrier crater, obtaining position information of the target barrier crater; based on the position information, determining a remote sensing image to which the target barrier crater belongs, and matching a digital elevation model homologous to the remote sensing image; based on the remote sensing image and the digital elevation model, segmenting a contour area of ​​the target barrier crater and its corresponding sputtering in the remote sensing image; determining geometric feature parameters of the contour area, and based on the geometric feature parameters, evaluating the morphological features of the target barrier crater.

[0096] Through the technical solution of the present invention, the present invention obtains the position information of the target barrier crater by responding to the morphological feature evaluation signal of the target barrier crater; and based on the position information, determines the remote sensing image to which the target barrier crater belongs, and matches the digital elevation model homologous to the remote sensing image; then, based on the remote sensing image and the digital elevation model, segments the contour area of ​​the target barrier crater and its corresponding sputtering in the remote sensing image; finally determines the geometric feature parameters of the contour area, and evaluates the morphological features of the target barrier crater based on the geometric feature parameters. Thus, through the remote sensing image and the digital elevation model to which the target barrier crater belongs, the contour area of ​​the target barrier crater and its corresponding sputtering is segmented in the remote sensing image, and finally, according to the geometric feature parameters of the contour area, the morphological features of the target barrier crater are evaluated, which can realize the automatic evaluation of the morphological features of the barrier crater and avoid manual participation, thereby improving the evaluation efficiency and accuracy of the morphological features of the barrier crater on Mars.

[0097] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0098] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the morphological characteristics of a Martian rampart impact crater, characterized in that: include: In response to a morphological feature evaluation signal of a target barrier impact crater, obtaining position information of the target barrier impact crater; Based on the location information, determining the remote sensing image to which the target barrier crater belongs, and matching a digital elevation model homologous to the remote sensing image; Based on the remote sensing image and the digital elevation model, segmenting the target barrier impact crater and the contour area of ​​the corresponding sputtering in the remote sensing image; The geometric characteristic parameters of the contour area are determined, and based on the geometric characteristic parameters, the morphological characteristics of the target barrier impact crater are evaluated.

2. The method according to claim 1, characterized in that The location information is the latitude and longitude coordinates of the target barrier impact crater; Determining the remote sensing image to which the target barrier impact crater belongs based on the position information includes: Converting the longitude and latitude coordinates of the target barrier crater to the coordinates of the Martian sphere where the target barrier crater is located to obtain the converted coordinates of the target barrier crater; Based on the converted coordinates, the map range of the target barrier crater on the Martian sphere is determined, and a remote sensing image matching the map range is searched in a preset remote sensing image database, wherein the preset remote sensing image database stores remote sensing images corresponding to different map ranges.

3. The method according to claim 1, characterized in that The matching of the digital elevation model homologous to the remote sensing image comprises: Based on the position information of the target barrier crater, determining an initial digital elevation model corresponding to the target barrier crater; Projecting the remote sensing image and the initial digital elevation model onto the same coordinate system, and processing the projected remote sensing image and the projected initial digital elevation model to have the same resolution, thereby obtaining the processed remote sensing image and the processed initial digital elevation model; The feature points of the target barrier crater are matched between the processed remote sensing image and the processed initial digital elevation model, and based on the matching result, a digital elevation model homologous to the remote sensing image is identified in the initial digital elevation model.

4. The method according to claim 1, characterized in that: The step of segmenting the target barrier impact crater and its corresponding sputtering object contour area in the remote sensing image based on the remote sensing image and the digital elevation model includes: Obtaining prompt information for segmenting the remote sensing image; Acquire a remote sensing large model, wherein the remote sensing large model includes a multi-head self-attention mechanism for extracting impact crater features, a feedforward neural network for transforming impact crater features, an encoder for encoding prompt information, and an image segmentation network for segmenting remote sensing images; The remote sensing image, the digital elevation model, and the prompt information are inputted into the remote sensing large model together, the crater features are extracted from the remote sensing image and the digital elevation model by the multi-head self-attention mechanism to obtain crater features, the crater features are nonlinearly transformed by the feedforward neural network to obtain nonlinear crater features, the prompt information is encoded by the encoder to obtain encoding features, the nonlinear crater features and the encoding features are image segmented by the image segmentation network to obtain the contour area of ​​the target barrier crater and its corresponding sputtering object; After segmenting the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model, the method further includes: If the contour area is a label mask file, the label mask file is converted into a vector file containing the geometric feature parameters of the contour area using a preset conversion method, wherein the method of converting the label mask file into a vector file containing the geometric feature parameters of the contour area using a preset conversion method includes: Converting the label mask file into a binary image, and processing the binary image into a raster image; Determining a geographic transformation matrix of the raster image, and extracting an affine matrix for transforming from pixel coordinates to geographic coordinates from the geographic transformation matrix; The inverse matrix of the affine matrix is ​​calculated, and based on the inverse matrix, the Martian geographic coordinates of the contour area are converted into pixel coordinates, and based on the contour area after the pixel coordinates are converted, a vector file containing the geometric feature parameters of the contour area is determined.

5. The method according to claim 1, characterized in that: The geometric characteristic parameters include the pit area and pit radius of the target barrier impact pit in the contour area, and the sputtering perimeter and sputtering area of ​​the sputtering seat formed by each of the sputtering objects; The step of evaluating the morphological characteristics of the target barrier crater based on the geometric characteristic parameters includes: Determining a flow coefficient of the sputtered material corresponding to the target barrier impact crater based on the crater area, the sputtering area, and the crater radius; Determining a lobation factor of the target barrier impact crater based on the sputtering perimeter and the sputtering area; Based on the flow coefficient and the lobation coefficient, the morphological characteristics of the target barrier impact crater are determined.

6. The method according to claim 1, characterized in that Before matching the digital elevation model homologous to the remote sensing image, the method further comprises: Filtering the remote sensing image using a preset filter, and projecting the filtered remote sensing image to the Martian geographic coordinate system where the target barrier impact crater is located, to obtain the projected remote sensing image; Determining the center position and the extension range of the target barrier crater, and based on the center position and the extension range, cutting out a target remote sensing image corresponding to the target barrier crater from the projected remote sensing image; The matching of the digital elevation model homologous to the remote sensing image comprises: Matching a digital elevation model that is homologous to the target remote sensing image.

7. The method according to claim 1, characterized in that After segmenting the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model, the method further includes: Taking the contour area as the contour area to be verified, obtaining the reference contour area of ​​the target barrier impact crater and its corresponding sputtering product segmented in the remote sensing image, and determining the reference contour area and reference contour perimeter of the reference contour area, and the contour area to be verified and the contour perimeter to be verified of the contour area to be verified; Calculating the area overlap between the reference contour area and the contour area to be verified, and determining the perimeter detection parameter based on the reference contour perimeter and the contour perimeter to be verified, and determining the area detection parameter based on the reference contour area and the contour area to be verified; Based on the region overlap, the perimeter detection parameters, and the area detection parameters, determine whether the contour area to be verified meets the segmentation requirements. If so, determine the morphological features of the target barrier impact crater based on the contour area to be verified. Otherwise, re-segment the contour area of ​​the target barrier impact crater and its corresponding spatter in the remote sensing image.

8. A device for evaluating the morphological characteristics of a Martian rampart impact crater, characterized in that: include: an acquisition unit, configured to acquire position information of the target barrier impact crater in response to a morphological feature evaluation signal of the target barrier impact crater; A matching unit, configured to determine the remote sensing image to which the target barrier crater belongs based on the position information, and to match a digital elevation model homologous to the remote sensing image; A segmentation unit, configured to segment the target barrier impact crater and its corresponding sputtering product contour area in the remote sensing image based on the remote sensing image and the digital elevation model; An evaluation unit is used to determine geometric characteristic parameters of the contour area and evaluate the morphological characteristics of the target barrier crater based on the geometric characteristic parameters.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images