A method for tracing the origin of rocks in lunar rover exploration areas using millimeter-level scene reconstruction

Abstract

This invention discloses a method for tracing the origin of rocks in the lunar rover exploration area using millimeter-level scene reconstruction, belonging to the field of lunar surface rock origin tracing technology. The method includes: S1, VNIS spectral data preprocessing; S2, spectral feature similarity analysis; S3, 3D scene reconstruction data preparation; S4, sparse matching and dense point cloud generation; S5, generation of DEM and DOM; S6, rock origin direction determination; S7, rock origin direction grouping and fitting; and S8, comprehensive rock origin analysis. Based on spectral information tracing, this invention introduces high-resolution visible light imagery from PCAM, proposing a new method for tracing the origin of materials in the Yutu-2 rover exploration area using millimeter-level 3D scenes. This method provides more evidence for rock origin tracing, achieves millimeter-level scene reconstruction of the Yutu-2 exploration area, and can be used to analyze the contact between rocks and lunar regolith, determine the rock origin direction, and thus provide richer evidence for lunar surface rock origin tracing, improving the reliability of the tracing results.

Description

Technical Field

[0001] This invention relates to the field of lunar surface rock tracing technology, specifically to a method for tracing the origin of rocks in a lunar rover exploration area using millimeter-level scene reconstruction. Background Technology

[0002] The Moon, Earth's only natural satellite and the closest extraterrestrial body, possesses unique geographical advantages, distinctive environmental resources, and abundant material resources. It has become the primary target for human space exploration and development, attracting widespread attention and research since the invention of the telescope in the 17th century and remaining a hot topic in international deep space exploration. China's lunar exploration mission is proceeding according to plan. Chang'e-4 successfully landed in the Von Kármán crater within the South Pole-Aitken (SPA) basin on the far side of the Moon on January 3, 2019. Its onboard rover (Yutu-2) conducted lunar exploration missions. Yutu-2 carries a panoramic camera and an infrared imaging spectrometer (VNIS) for near-field observation. The VNIS includes CMOS and SWIR sensors, capable of acquiring spectral data in the 450nm-950nm and 900nm-2400nm bands, respectively. These two payloads can obtain high-resolution, high-precision morphological and spectral data of the lunar surface, providing reliable data for lunar rock detection. In particular, small rocks that cannot be observed by orbiters can be clearly displayed by Yutu-2's observation equipment. Small rocks provide important material transport information representing key events such as the origin and evolution of lunar soil and meteorite impacts, which is of great research value for the study of geological evolution history. Therefore, this invention provides a method for tracing the origin of rocks in lunar rover exploration areas through millimeter-level scene reconstruction.

[0003] The existing technology has the following problems:

[0004] 1. Existing methods for tracing the origin of rocks in lunar rover exploration areas using millimeter-level scene reconstruction mostly rely on spectral analysis, which interprets the origin of rocks based on the similarity of their material composition. However, this method has certain limitations. The spectral characteristics of lunar rocks sometimes differ very little, making it difficult to provide more detailed evidence of the rock's origin. Therefore, it is not sufficient to determine that these rocks share the same origin. Summary of the Invention

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for tracing the origin of rocks in a lunar rover's exploration area through millimeter-level scene reconstruction includes the following steps:

[0007] S1. VNIS spectral data preprocessing: First, the spectral radiance data of the Yutu-2 VNIS sensor is converted into bidirectional reflectance using the solar irradiance calibration method; second, an offset is added to the CMOS segment data to connect it to the SWIR segment data; third, the spectral reflectance data is smoothed to eliminate possible noise and discontinuities; finally, the spectral reflectance data is processed using the continuum removal method to highlight spectral features.

[0008] S2. Spectral Feature Similarity Analysis: The wavelength values ​​of the absorption centers near 1000 nm and 2000 nm in the spectral data after continuum removal are calculated. The wavelength values ​​of the absorption centers near 1000 nm are plotted on the x-axis and the wavelength values ​​of the absorption centers near 2000 nm are plotted on the y-axis. By calculating the Euclidean distance between the absorption center wavelength values ​​of the rock spectral data and the absorption center wavelength values ​​of known typical impact crater regions, the degree of similarity is judged. The smaller the value, the more similar the similarity. This can explain the similarity of the composition of the rock and the typical impact crater region, and thus provide preliminary evidence of the rock's origin.

[0009] S3. Preparation of 3D scene reconstruction data: 3D scene reconstruction requires high-quality Yutu-2 PCAM images. First, the quality of the images is judged based on their visual characteristics. The image influencing factors considered in the judgment include blur, overexposure, underexposure, artifacts, etc. By comprehensively considering these factors, those images judged to be of poor quality will be deleted from the dataset to avoid these images from having an adverse effect on subsequent analysis and research.

[0010] S4. Sparse Matching and Dense Point Cloud Generation: First, the scale-invariant feature transform algorithm is used to extract and match feature points in the image. Through feature matching, corresponding points between images are determined, and their correspondence is calculated. Then, the structure for motion reconstruction is applied to calculate the interior and exterior orientation elements of each image, thereby establishing the geometric relationship between the images. After calculating the interior and exterior orientation elements, a multi-view stereo matching algorithm is applied to obtain a dense point cloud model composed of a series of three-dimensional points. Before sparse matching, the camera calibration parameters are set with a pixel size of 7.4mm × 7.4mm and a focal length of 50.274mm.

[0011] S5. Generate DEM and DOM: Based on the dense point cloud, generate the DEM of the study area through interpolation; based on the generated DEM and the corresponding PCAM image, perform geometric correction, and project the pixels in the image onto the corresponding positions on the ground to obtain the DOM of the study area.

[0012] S6. Determining the direction of rock origin: Preliminary determination of the direction of rock origin: Observe the accumulation of lunar regolith around the rocks in the study area from PCAM images, 3D point clouds, and DOM, and visually identify the degree of uplift at the contact point between the lunar regolith and the rocks. The opposite direction of the side with higher uplift is the direction of rock origin. Secondly, calculate the topographic slope map in the preliminary direction of rock origin using DEM, and then calculate the local slope of the lunar regolith. The side with the smaller absolute value of the slope is the direction of rock origin.

[0013] S7. Grouping and Fitting Rock Origin Direction: Group a series of rocks that are close in location and have roughly the same origin direction into a group, thereby dividing the rocks in the study area into several groups. Then, measure the coordinates of the rocks in an independent coordinate system, and fit the coordinates of each group of rocks into a linear expression based on the least squares fitting principle, thus obtaining the specific origin direction of the rocks.

[0014] S8. Comprehensive analysis of rock origin: Taking into account the evidence of rock composition similarity provided by spectral data and the evidence of rock origin direction provided by the three-dimensional scene of the study area, if the rock origin direction basically points to the direction of the impact crater with similar rock composition, then the rock should be the rock ejected from that impact crater. If there is a large difference, it indicates that the rock may have been ejected to that place due to multiple impact events, thus realizing the analysis of the rock origin and evolution process in the Yutu-2 exploration area.

[0015] Preferably, the bidirectional reflectivity conversion BRF of S1 is π·I·D 2 / E0(λ)·cos(θ), where BRF is the bidirectional reflectance, I is the VNIS radiance data; E0(λ) represents the solar irradiance at a distance of 1 astronomical unit; D is the real-time solar-moon distance at the time of observation, in AU; θ is the solar zenith angle. The offset added to the CMOS segment data is the difference between the average bidirectional reflectance of the CMOS data and the SWIR data in the overlapping band from 900nm to 945nm. This offset is added to the CMOS data from 450 to 895nm, while the spectral reflectance data from 900nm to 2395nm completely retains the SWIR segment data.

[0016] Preferably, data smoothing: an 11-point sliding smoothing algorithm is used to reduce noise and discontinuities in the data; continuum removal: the continuum range is 800-2300nm, and it is a linear continuum; the continuum removal formula... R cj It is the spectrum with the continuous spectrum removed from the J-band, R j It is the original spectral value of band j, λ j It is the wavelength of the J-band, R start and R end These are the spectral values ​​of the starting and ending points, respectively, λ. start and λ endIt refers to the wavelengths at the start and end points.

[0017] Preferably, S2 uses the minimum values ​​of the bidirectional reflectance data around 1000nm and 2000nm as the center, and selects three points on each side for a total of seven points. A parabola is fitted to these points, and the minimum value of the parabola equation is taken as the wavelength value of the absorption center. The absorption center similarity judgment method... The values ​​represent the wavelengths of the absorption centers around 1000 nm for reflectance data from rocks and typical impact crater regions, respectively. The values ​​are the wavelengths of the absorption centers around 2000 nm for reflectance data of rocks and typical impact crater regions, respectively. The smaller the value of D, the higher the detail of the composition of the rocks and typical impact crater regions, thus indicating that the rocks originated from the impact crater.

[0018] Preferably, in determining the direction of rock origin, S6 qualitatively determines the direction of rock origin by judging the degree of accumulation of lunar weathering material at the contact point with the rock. The side with more accumulation is due to the compression effect during rock ejection, which causes the local uplift of the lunar weathering layer. In quantitatively determining the direction of rock origin, the absolute value of the local slope of the weathering material accumulation in the DEM slope is calculated. The slope calculation range is: starting from the rock boundary, extending to a short distance where the profile line shows a relatively obvious inflection point. The direction of rock origin is the side with the smaller slope.

[0019] Preferably, S7 establishes an independent Cartesian coordinate system with the PCAM camera's imaging center as the origin, the lunar north pole direction as the Y-axis, and the east direction as the X-axis. The coordinates (Y, X) of each rock after grouping are measured from the DOM in the independent coordinate system. Under the constraint of the least squares criterion, the source direction is fitted into a linear expression of X = P1 * Y + P2, thereby providing the source direction of the rock.

[0020] Due to the adoption of the above technical solution, the technical progress achieved by this invention compared to the prior art is as follows:

[0021] 1. This invention provides a method for tracing the origin of rocks in the lunar rover exploration area using millimeter-level scene reconstruction. Addressing the problem that traditional methods relying on spectral data for material composition tracing often result in contradictory or incomplete tracing results due to small differences in spectral characteristics, this invention introduces high-resolution visible light image data from the Yutu-2 PCAM rover, proposing a novel method for tracing the origin of materials in the Yutu-2 exploration area using millimeter-level three-dimensional scenes. This method can provide more evidence for rock origin tracing.

[0022] 2. This invention provides a method for tracing the origin of rocks in a lunar rover exploration area through millimeter-level scene reconstruction. The constructed millimeter-level three-dimensional scene can accurately determine the contact relationship between rocks and lunar surface weathering materials, quantitatively determine the slope of lunar surface weathering materials, and thus provide evidence of the direction of rock origin. By fitting the origin directions of several rocks with similar origin directions into a linear expression, combined with spectral origin evidence, it helps to further interpret the evolution process of the Yutu-2 exploration area and improve the lunar formation and evolution model. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall process structure of a method for tracing the source of rocks in a lunar rover exploration area using millimeter-level scene reconstruction according to the present invention. Detailed Implementation

[0024] The present invention will be further described in detail below with reference to embodiments:

[0025] like Figure 1 As shown, this invention provides a method for tracing the origin of rocks in a lunar rover exploration area through millimeter-level scene reconstruction, comprising the following steps:

[0026] S1. VNIS spectral data preprocessing: First, the spectral radiance data of the Yutu-2 VNIS sensor is converted into bidirectional reflectance using the solar irradiance calibration method; second, an offset is added to the CMOS segment data to connect it to the SWIR segment data; third, the spectral reflectance data is smoothed to eliminate possible noise and discontinuities; finally, the spectral reflectance data is processed using the continuum removal method to highlight spectral features.

[0027] S2. Spectral Feature Similarity Analysis: The wavelength values ​​of the absorption centers near 1000 nm and 2000 nm in the spectral data after continuum removal are calculated. The wavelength values ​​of the absorption centers near 1000 nm are plotted on the x-axis and the wavelength values ​​of the absorption centers near 2000 nm are plotted on the y-axis. By calculating the Euclidean distance between the absorption center wavelength values ​​of the rock spectral data and the absorption center wavelength values ​​of known typical impact crater regions, the degree of similarity is judged. The smaller the value, the more similar the similarity. This can explain the similarity of the composition of the rock and the typical impact crater region, and thus provide preliminary evidence of the rock's origin.

[0028] S3. Preparation of 3D scene reconstruction data: 3D scene reconstruction requires high-quality Yutu-2 PCAM images. First, the quality of the images is judged based on their visual characteristics. The image influencing factors considered in the judgment include blur, overexposure, underexposure, artifacts, etc. By comprehensively considering these factors, those images judged to be of poor quality will be deleted from the dataset to avoid these images from having an adverse effect on subsequent analysis and research.

[0029] S4. Sparse Matching and Dense Point Cloud Generation: First, the scale-invariant feature transform algorithm is used to extract and match feature points in the image. Through feature matching, corresponding points between images are determined and their correspondence is calculated. Then, the structure of motion reconstruction is applied to calculate the interior and exterior orientation elements of each image, thereby establishing the geometric relationship between the images. After calculating the interior and exterior orientation elements, the multi-view stereo matching algorithm is applied to obtain a dense point cloud model composed of a series of three-dimensional points. Before sparse matching, the pixel size in the camera calibration parameter settings is 7.4mm×7.4mm and the focal length is 50.274mm.

[0030] S5. Generate DEM and DOM: Based on the dense point cloud, generate the DEM of the study area through interpolation; based on the generated DEM and the corresponding PCAM image, perform geometric correction, and project the pixels in the image onto the corresponding positions on the ground to obtain the DOM of the study area.

[0031] S6. Determining the direction of rock origin: Preliminary determination of the direction of rock origin: Observe the accumulation of lunar regolith around the rocks in the study area from PCAM images, 3D point clouds, and DOM, and visually identify the degree of uplift at the contact point between the lunar regolith and the rocks. The opposite direction of the side with higher uplift is the direction of rock origin. Secondly, calculate the topographic slope map in the preliminary direction of rock origin using DEM, and then calculate the local slope of the lunar regolith. The side with the smaller absolute value of the slope is the direction of rock origin.

[0032] S7. Grouping and Fitting Rock Origin Direction: Group a series of rocks that are close in location and have roughly the same origin direction into a group, thereby dividing the rocks in the study area into several groups. Then, measure the coordinates of the rocks in an independent coordinate system, and fit the coordinates of each group of rocks into a linear expression based on the least squares fitting principle, thus obtaining the specific origin direction of the rocks.

[0033] S8. Comprehensive analysis of rock origin: Taking into account the evidence of rock composition similarity provided by spectral data and the evidence of rock origin direction provided by the three-dimensional scene of the study area, if the rock origin direction basically points to the direction of the impact crater with similar rock composition, then the rock should be the rock ejected from that impact crater. If there is a large difference, it indicates that the rock may have been ejected to that place due to multiple impact events, thus realizing the analysis of the rock origin and evolution process in the Yutu-2 exploration area.

[0034] In this implementation case, before sparse matching, the camera's calibration parameters are set with a pixel size of 7.4mm × 7.4mm and a focal length of 50.274mm. The sparse matching uses the Align Photos module in Agisoft Metashape software, with accuracy set to High. Generic preselection and Reference preselection are both checked by default, and the Keypoint limit parameter and Tie point limit parameter are set to the default value of 40000. To generate dense point clouds, the Build Dense Cloud module in Agisoft Metashape software was called with Quality set to High and Depth filtering to Mild. Calculate point colors was selected. When generating DEM and DOM, the Build DEM module in Agisoft Metashape software was called first with the parameters set to default. After generating the DEM, the Build Orthomosaic module was called to generate the DOM with the parameters set to default. To address the problem that traditional methods of tracing the origin of materials based on spectral data often result in contradictory or incomplete results due to small differences in spectral features, a new method for tracing the origin of materials in the Yutu-2 exploration area using millimeter-level 3D scenes was proposed, based on spectral data tracing. This method can provide more evidence for rock origin tracing.

[0035] like Figure 1 As shown, the present invention provides a technical solution: preferably, S1 bidirectional reflectivity conversion BRF = π·I·D 2 / E0(λ)·cos(θ), where BRF is the bidirectional reflectance, I is the VNIS radiance data; E0(λ) represents the solar irradiance at a distance of 1 astronomical unit; D is the real-time solar-moon distance at the time of observation, in AU; θ is the solar zenith angle. The offset added to the CMOS segment data is the difference in average bidirectional reflectance between the CMOS data and the SWIR data in the overlapping band from 900nm to 945nm. This offset is added to the CMOS data from 450 to 895nm, while the spectral reflectance data from 900nm to 2395nm is completely preserved from the SWIR segment data. Data smoothing: An 11-point sliding smoothing algorithm is used to reduce noise and discontinuities in the data. Continuum removal: The continuum range is 800-2300nm, which is a linear continuum. The continuum removal formula is... R cj It is the spectrum with the continuous spectrum removed from the J-band, R j It is the original spectral value of band j, λ j It is the wavelength of the J-band, Rstart and R end These are the spectral values ​​of the starting and ending points, respectively, λ. start and λ end The wavelengths at the start and end points are used. S2 is centered on the minimum values ​​of the bidirectional reflectance data around 1000nm and 2000nm, with three points selected on each side for a total of seven points. A parabola is fitted, and the minimum value of the parabola equation is taken as the wavelength value of the absorption center. The absorption center similarity judgment method is then used. The values ​​represent the wavelengths of the absorption centers around 1000 nm for reflectance data from rocks and typical impact crater regions, respectively. The values ​​represent the wavelengths of the absorption centers around 2000 nm for reflectance data from rocks and typical impact crater regions. A smaller D value indicates a higher level of detail in the composition of the rocks and typical impact crater regions, suggesting the rocks originated from that impact crater. S6 qualitatively determines the rock origin direction by assessing the degree of accumulation of lunar regolith at the contact point with the rock. The side with more accumulation is attributed to the compression effect of rock ejection, resulting in a local uplift of the lunar regolith. Quantitatively determining the rock origin direction involves calculating the local accumulation of weathering material on the DEM slope. The absolute value of the slope is calculated within the range of the rock boundary, extending to a short distance from which a noticeable inflection point appears on the profile line. The direction of rock origin is the side with the smaller slope. S7 establishes an independent Cartesian coordinate system with the PCAM camera's imaging center as the origin, the lunar north pole direction as the Y-axis, and the east direction as the X-axis. The coordinates (Y, X) of each rock in the independent coordinate system after grouping are measured from the DOM. Under the constraint of the least squares criterion, the direction of origin is fitted into a linear expression of X = P1 * Y + P2, thus providing the direction of rock origin.

[0036] In this embodiment, the constructed millimeter-level three-dimensional scene can accurately determine the contact relationship between rocks and lunar surface weathering materials, quantitatively determine the slope of lunar surface weathering materials, and thus provide evidence of the direction of rock origin. By fitting the origin directions of several rocks with similar origin directions into the form of a linear expression, combined with spectral source tracing evidence, it helps to further interpret the evolution process of the Yutu-2 exploration area and improve the lunar formation and evolution model.

[0037] The working principle of this method for tracing the origin of rocks in the lunar rover's exploration area through millimeter-level scene reconstruction will be explained in detail below.

[0038] like Figure 1As shown, the method for tracing the origin of rocks in the Yutu-2 rover exploration area mainly involves combining in-situ spectral data collected by the rover with large-scale spectral data collected by the orbiter. By judging the similarity of absorption center wavelengths in the spectral absorption characteristics, the similarity of material composition is inferred, thus conducting rock origin analysis. Current research indicates that the material in the Yutu-2 exploration area mainly originates from the Finsen crater, but some researchers suggest that some material originated from the Vega crater, which remains highly controversial. Simply analyzing the similarity of material composition through spectral analysis is insufficient to resolve these contradictions. Rocks originating from impact events typically contact the lunar regolith at a certain angle. The lunar regolith is very soft, causing it to accumulate on one side of the rock. Based on this phenomenon, this method constructs a millimeter-level 3D scene of the Yutu-2 exploration area using high-resolution PCAM images from Chang'e-2, based on the principle of 3D reconstruction. This allows for precise observation and quantitative description of the contact relationship between rocks and the regolith, thereby determining the rock's origin. Linear fitting is performed on a series of rocks with roughly the same origin direction to avoid randomness.

[0039] The present invention has been described in detail above. However, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, any modifications or improvements that do not depart from the spirit of the present invention are within the scope of protection of the present invention.

Claims

1. A method for tracing the origin of rocks in a lunar rover exploration area through millimeter-level scene reconstruction, comprising the following steps, characterized in that: S1. VNIS spectral data preprocessing: First, the spectral radiance data of the Yutu-2 VNIS sensor is converted into bidirectional reflectance using the solar irradiance calibration method; second, an offset is added to the CMOS segment data to connect it to the SWIR segment data; third, the spectral reflectance data is smoothed to eliminate possible noise and discontinuities; finally, the spectral reflectance data is processed using the continuum removal method to highlight spectral features. S2. Spectral Feature Similarity Analysis: The wavelength values ​​of the absorption centers near 1000 nm and 2000 nm in the spectral data after continuum removal are calculated. The wavelength values ​​of the absorption centers near 1000 nm are plotted on the x-axis and the wavelength values ​​of the absorption centers near 2000 nm are plotted on the y-axis. By calculating the Euclidean distance between the absorption center wavelength values ​​of the rock spectral data and the absorption center wavelength values ​​of known typical impact crater regions, the degree of similarity is judged. The smaller the value, the more similar the similarity. This can explain the similarity of the composition of the rock and the typical impact crater region, and thus provide preliminary evidence of the rock's origin. S3. Preparation of 3D scene reconstruction data: 3D scene reconstruction requires high-quality Yutu-2 PCAM images. First, the quality of the images is judged based on their visual characteristics. The image influencing factors considered in the judgment include blur, overexposure, underexposure, and artifacts. By comprehensively considering these factors, those images judged to be of poor quality will be deleted from the dataset to avoid these images from having an adverse effect on subsequent analysis and research. S4. Sparse Matching and Dense Point Cloud Generation: First, the scale-invariant feature transform algorithm is used to extract and match feature points in the image. Through feature matching, corresponding points between images are determined and their correspondence is calculated. Then, the structure for motion reconstruction is applied to calculate the interior and exterior orientation elements of each image, thereby establishing the geometric relationship between the images. After calculating the interior and exterior orientation elements, the multi-view stereo matching algorithm is applied to obtain a dense point cloud model composed of a series of three-dimensional points. Before sparse matching, the camera calibration parameters are set with a pixel size of 7.4mm×7.4mm and a focal length of 50.274mm. S5. Generate DEM and DOM: Based on the dense point cloud, generate the DEM of the study area through interpolation; based on the generated DEM and the corresponding PCAM image, perform geometric correction, and project the pixels in the image onto the corresponding positions on the ground to obtain the DOM of the study area. S6. Determining the direction of rock origin: Preliminary determination of the direction of rock origin: Observe the accumulation of lunar regolith around the rocks in the study area from PCAM images, 3D point clouds, and DOM. Visually identify the degree of uplift at the contact point between the lunar regolith and the rocks. The opposite direction of the side with higher uplift is the direction of rock origin. Secondly, calculate the topographic slope map in the preliminary direction of rock origin using DEM, and then calculate the local slope of the lunar regolith. The side with the smaller absolute value of the slope is the direction of rock origin. S7. Grouping and Fitting Rock Origin Direction: Group a series of rocks that are close in location and have roughly the same origin direction into a group, thereby dividing the rocks in the study area into several groups. Then, measure the coordinates of the rocks in an independent coordinate system, and fit the coordinates of each group of rocks into a linear expression based on the least squares fitting principle, thus obtaining the specific origin direction of the rocks. S8. Comprehensive analysis of rock origin: Taking into account the evidence of rock composition similarity provided by spectral data and the evidence of rock origin direction provided by the three-dimensional scene of the study area, if the rock origin direction basically points to the direction of the impact crater with similar rock composition, then the rock should be the rock ejected from that impact crater. If there is a large difference, it indicates that the rock may have been ejected to that place due to multiple impact events, thus realizing the analysis of the rock origin and evolution process in the Yutu-2 exploration area.

2. The method for tracing the origin of rocks in a lunar rover exploration area using millimeter-level scene reconstruction according to claim 1, characterized in that: S1 bidirectional reflectivity conversion Where BRF is bidirectional reflectance and I is VNIS radiance data; Represents the solar irradiance at a distance of 1 astronomical unit (AU); D is the real-time distance between the sun and the moon at the time of observation, in AU. The offset added to the CMOS segment data is the difference in average bidirectional reflectance between the CMOS data and the SWIR data in the overlapping band from 900nm to 945nm, which is the solar zenith angle. This offset is added to the CMOS data from 450nm to 895nm, while the spectral reflectance data from 900nm to 2395nm is completely preserved from the SWIR segment data.

3. The method for tracing the origin of rocks in a lunar rover exploration area through millimeter-level scene reconstruction according to claim 2, characterized in that: Data smoothing: An 11-point sliding smoothing algorithm is used to reduce noise and discontinuities in the data. Continuum removal: The continuous series ranges from 800-2300 nm and is a linear continuous series. The formula for continuous series removal is... It is the spectrum after removing the continuous spectrum from the J-band. These are the original spectral values ​​for band J. It is the J-band wavelength. and These are the spectral values ​​of the starting point and the ending point, respectively. and It refers to the wavelengths at the start and end points.

4. The method for tracing the origin of rocks in a lunar rover exploration area using millimeter-level scene reconstruction according to claim 1, characterized in that: S2 uses the minimum values ​​of bidirectional reflectance data around 1000nm and 2000nm as centers, and selects three points on each side for a total of seven points. A parabola is fitted, and the minimum value of this parabola equation is taken as the wavelength value of the absorption center. The absorption center similarity judgment method... The values ​​represent the wavelengths of the absorption centers around 1000 nm for reflectance data from rocks and typical impact crater regions, respectively. The values ​​are the wavelengths of the absorption centers around 2000 nm for reflectance data of rocks and typical impact crater regions, respectively. The smaller the value of D, the higher the detail of the composition of the rocks and typical impact crater regions, thus indicating that the rocks originated from the impact crater.

5. The method for tracing the origin of rocks in a lunar rover exploration area using millimeter-level scene reconstruction according to claim 1, characterized in that: When determining the direction of rock origin, S6 qualitatively determines the direction of rock origin by judging the degree of accumulation of lunar weathering material at the contact point with the rock. The side with more accumulation is due to the compression effect of rock ejection, which causes the local uplift of the lunar weathering layer. When quantitatively determining the direction of rock origin, the absolute value of the local slope of the weathering material accumulation in the DEM slope is calculated. The slope calculation range is: starting from the rock boundary, extending to a short distance where the profile line has a more obvious inflection point. The direction of rock origin is the side with the smaller slope.

6. The method for tracing the origin of rocks in a lunar rover exploration area using millimeter-level scene reconstruction according to claim 1, characterized in that: S7 establishes an independent Cartesian coordinate system with the PCAM camera's imaging center as the origin, the lunar north pole as the Y-axis, and the east direction as the X-axis. The coordinates (Y, X) of each grouped rock in this independent coordinate system are measured from the DOM. Under the constraint of the least squares criterion, the source direction is fitted as... The linear expression is used to obtain the direction of rock origin.

Images