A method for generating a ground elevation model based on co-orbit satellites

Abstract

The present invention discloses a method for generating a ground elevation model based on co-orbiting satellites, which relates to the field of Internet technology. By collecting elevation data at multiple time periods and performing preprocessing to form an elevation data set D, temporal consistency detection and correction are performed on the elevation data set D based on a consistency factor C to obtain a corrected elevation data set Dc. The short-term elevation change index △H is calculated to mark and eliminate dynamic interference points, and the elevation data set Ds after elimination is obtained. Based on the calculation of the elevation fusion evaluation index Hf, multi-time period data is integrated to form a ground elevation data set DEMb, and by comparing with a preset accuracy allowable error threshold Tthe, the available state result of the ground elevation data set DEMb is obtained. Through iterative optimization, the construction of a high-precision ground elevation model is completed, realizing the efficient fusion of multi-time period elevation data after dynamically eliminating interference points, and generating a stable ground elevation model that can reflect dynamic change characteristics.

Description

Technical Field

[0001] The present invention relates to the field of Internet technology, and specifically to a method for generating a ground elevation model based on co-orbiting satellites. Background Art

[0002] The ground elevation model is one of the core research directions in the field of geoinformation science, and it has important application values in disaster prediction, urban planning, resource management, etc. As a specific research direction in this field, using satellite remote sensing technology to construct an elevation model, with its advantages of wide coverage and high observation efficiency, has become one of the important means for high-precision topographic mapping. In satellite remote sensing technology, the application of co-orbiting satellites is particularly unique. It refers to satellites with the same orbital parameters, which can repeatedly observe the same surface area in a periodic manner. This characteristic makes it possible to obtain high-precision topographic data with multiple time series.

[0003] In actual application scenarios, such as the assessment of terrain changes in flood monitoring, the rapid restoration modeling of the terrain in the disaster area after an earthquake, and the micro-topography analysis of crop planting areas, multiple observations by co-orbiting satellites can provide key data for accurate and efficient terrain modeling.

[0004] Although co-orbiting satellites have significant advantages in generating ground elevation models, the existing processing methods have obvious limitations in multi-source data fusion. Specifically, due to the data generated by multiple flights being affected by observation time, weather conditions, and surface dynamic changes, such as rainfall and seasonal vegetation growth factors, the quality and consistency differences between different data sets are relatively large, resulting in accuracy deviations in the generated elevation models. In addition, the current data processing algorithms usually rely on simple averaging or interpolation methods, and cannot effectively eliminate the interference of temporary surface changes on the model. This deficiency not only limits the application potential of the elevation model in dynamic and complex terrains, but also reduces the reliability of rapid response in emergency scenarios such as natural disasters. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method for generating a ground elevation model based on co-orbiting satellites, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for generating a ground elevation model based on co-orbiting satellites, comprising the following steps:

[0007] S1. Collect multi-period elevation data of co-orbiting satellites, obtain relevant observation data, and perform preprocessing to form an elevation data set D;

[0008] S2. Extract features from the elevation dataset D, then perform temporal consistency detection to determine the stability of the elevation dataset D. By quantifying the consistency of multi-temporal data, obtain the consistency factor C, and perform correction according to the consistency factor C to obtain the corrected elevation dataset Dc;

[0009] S3. Analyze the short-term surface changes of the corrected elevation dataset Dc, calculate the short-term elevation change index △H to dynamically eliminate interference points, and mark the interference points to obtain the eliminated elevation data set Ds after dynamic elimination;

[0010] S4. By fusing the eliminated elevation data sets Ds of different time series, obtain the elevation fusion evaluation index Hf, and perform integration to form the ground elevation dataset DEMb, and judge the comparison with the preset accuracy allowable error threshold Tthe to obtain the available status result of the ground elevation dataset DEMb;

[0011] S5. Perform iterative optimization and construct a ground elevation model according to the available status result of the ground elevation dataset DEMb.

[0012] Preferably, S1 includes S11 and S12;

[0013] S11. Collect multi-temporal elevation data of the same-track satellite, record the elevation data Hij of each observation and the relevant observation conditions, including the sampling time Ti and the illumination condition Li, and form the initial elevation data set Dr = {Hij, Ti, Li};

[0014] Among them, the elevation data Hij specifically represents the elevation value of the jth ground sampling point of the ith observation; the sampling time T specifically represents the observation timestamp of the ith sampling, which is used to identify the time sequence of the observation; the illumination condition Li represents the ambient illumination condition of the ith sampling.

[0015] Preferably, S12. Preprocess the initial elevation data set Dr, including noise elimination, outlier correction, and data standardization, to generate the elevation dataset D = {Hij(std), Ti, Li(std)}, where Hij(std) and Li(std) respectively represent the standardized elevation data and the standardized illumination condition after data standardization preprocessing;

[0016] Among them, noise elimination is performed by using the upper and lower quartile method to detect and eliminate noise points in the initial elevation data set Dr; outlier correction is performed by using the linear interpolation method to correct outliers in the initial elevation data set Dr after noise elimination; data standardization is performed by using the minimum-maximum normalization method to normalize the initial elevation data set Dr after outlier correction to the range of 0 to 1.

[0017] Preferably, S2 includes S21 and S22;

[0018] S21 extracts features from the elevation dataset D to obtain the standardized elevation data Hij(std) and the sampling time Ti, performs a temporal consistency check on the multi-temporal data of the same ground sampling point j, forms a time series feature set Hj = {H1j(std), H2j(std), H3j(std) ……, Hnj(std)} to judge the stability of the elevation dataset D, and obtains the consistency factor C by quantifying the consistency of the multi-temporal data;

[0019] Among them, n represents the total number of sampling observations, and Hj represents the elevation time series feature of the ground sampling point j;

[0020] The consistency factor C is obtained through the following calculation formula:

[0021]

[0022] In the formula, Cj represents the consistency factor at the position of the ground sampling point j, σ(Hj) represents the standard deviation of the time series feature set Hj, which is specifically used to measure the degree of data fluctuation, and Hmax and Hmin respectively represent the upper limit elevation value and the lower limit elevation value at the position of the ground sampling point j;

[0023] Among them, the standard deviation σ(Hj) is obtained through the calculation formula, and in the formula, represents the mean value of the standardized elevation data Hij(std) in the time series feature set.

[0024] Preferably, S22 compares the obtained consistency factor C with the preset consistency state threshold Cthe, obtains the result of the dataset correction execution state, and corrects the elevation dataset D according to the result of the dataset correction execution state to obtain the corrected elevation dataset Dc;

[0025] The result of the dataset correction execution state is obtained through the following comparison method:

[0026] When the consistency factor C < the consistency state threshold Cthe, the result of the dataset correction execution state is obtained as executing the correction, including correcting the standardized elevation data Hij(std) of the ground sampling point j to the mean value Hj, and adding the replaced data of the ground sampling point j to the corrected elevation dataset Dc;

[0027] When the consistency factor C ≥ the consistency state threshold Cthe, the result of the dataset correction execution state is obtained as not executing the correction, and the standardized elevation data Hij(std) of the ground sampling point j is directly added to the corrected elevation dataset Dc.

[0028] Preferably, S3 includes S31 and S32;

[0029] S31 analyzes short-term surface changes in the corrected elevation dataset Dc, including calculating the short-term fluctuations between adjacent time points for the same ground sampling point j, obtaining the short-term elevation change index ΔH, and dynamically removing interference points;

[0030] The short-term elevation change index ΔH is obtained through the following calculation formula:

[0031]

[0032] In the formula, ΔHij represents the short-term elevation change index at the position of the ground sampling point j in the i-th observation, Hij(std) represents the standardized elevation data at the position of the ground sampling point j in the corrected elevation dataset Dc for the i-th observation, H(i + 1)j(std) represents the standardized elevation data at the position of the ground sampling point j in the (i + 1)-th observation, and Ti and Tk represent adjacent sampling times, specifically representing the sampling time Ti and the sampling time T(i + 1).

[0033] Preferably, S32 compares the short-term elevation change index ΔHij with a preset short-term change evaluation threshold Hthe to determine the interference state result Mij at the position of the ground sampling point j, and marks the interference points at the position of the ground sampling point j according to the interference state result Mij, obtaining the removed elevation data set Ds after dynamically removing from the corrected elevation dataset Dc;

[0034] The interference state result Mij is obtained through the following comparison method:

[0035]

[0036] Among them, when the short-term elevation change index ΔHij > the short-term change evaluation threshold Hthe, the interference state result Mij = 1, specifically indicating that the position of the ground sampling point j needs to be marked as an interference point;

[0037] When the short-term elevation change index ΔHij ≤ the short-term change evaluation threshold Hthe, the interference state result Mij = 0, specifically indicating that the position of the ground sampling point j does not need to be marked as an interference point.

[0038] Preferably, S4 includes S41 and S42;

[0039] S41 fuses the removed elevation data sets Ds of different time series to obtain the elevation fusion evaluation index Hf, and integrates them to form the ground elevation dataset DEMb = {Hf, j, Xj, Yj}, where j represents the ground sampling point, and Xj and Yj represent the horizontal and vertical coordinates of the ground sampling point;

[0040] The elevation fusion evaluation index Hf is obtained through the following calculation formula:

[0041]

[0042] In the formula, Hij(std) represents the standardized elevation data of the j-th position of the ground sampling point in the i-th observation after removing the elevation data set Ds, and n represents the total number of sampling observations.

[0043] Preferably, in S42, the available state result of the ground elevation data set DEMb is obtained by extracting the comparison state with the preset accuracy allowable error threshold Tthe from the ground elevation data set DEMb;

[0044] The available state result of the ground elevation data set DEMb is obtained through the following comparison method:

[0045]

[0046] Among them, represents the existential logical symbol, represents the universal logical symbol;

[0047] When Mdem = 1, the available state result of the ground elevation data set DEMb is obtained as available;

[0048] When Mdem = 0, the available state result of the ground elevation data set DEMb is obtained as unavailable.

[0049] Preferably, the S5 includes S51;

[0050] S51. Trigger the execution of the iterative optimization mechanism according to the available state result of the ground elevation data set DEMb. When the available state result of the ground elevation data set DEMb is available, do not trigger the execution of the iterative optimization mechanism, and prompt to construct the ground elevation model. When the available state result of the ground elevation data set DEMb is unavailable, trigger the execution of the iterative optimization mechanism, including adjusting the accuracy allowable error threshold Tthe, the short-term change evaluation threshold Hthe, and the consistency state threshold Cthe.

[0051] The present invention provides a method for generating a ground elevation model based on in-orbit satellites, having the following beneficial effects:

[0052] (1) By collecting elevation data at multiple time periods and performing preprocessing to form an elevation data set D; performing temporal consistency detection and correction on the elevation data set D based on the consistency factor C to obtain the corrected elevation data set Dc; calculating the short-term elevation change index △H to mark and eliminate dynamic interference points to obtain the elevation data set Ds after elimination; integrating multi-period data based on the calculation of the elevation fusion evaluation index Hf to form the ground elevation data set DEMb, and comparing it with the preset accuracy allowable error threshold Tthe to obtain the available state result of the ground elevation data set DEMb; finally, completing the construction of a high-precision ground elevation model through iterative optimization, realizing the efficient fusion of multi-period elevation data after dynamically eliminating interference points, and generating a stable, high-precision ground elevation model that can reflect dynamic change characteristics. Compared with traditional methods, this scheme not only improves the accuracy and consistency of the ground elevation model, but also has strong robustness and adaptability, providing a scientific and efficient solution path for high-precision terrain modeling under complex terrain conditions.

[0053] (2) By quantifying the short-term elevation fluctuation degree of the ground sampling point j between adjacent time points, it effectively captures abnormal changes caused by temporary factors such as vegetation cover and rainwater ponding in a short period of time. Generate the interference state result

[0054] , by marking and dynamically eliminating interference points, generating the elevation data set Ds after elimination, thus maximizing the impact of dynamic interference points on elevation data, being able to accurately capture the elevation fluctuation of ground sampling points between adjacent time points, and also realizing precise dynamic elimination based on the short-term change evaluation threshold Hthe. Compared with traditional static elimination methods, it can dynamically mark and eliminate short-term abnormal points while retaining stable terrain change information, laying a more reliable dynamic change processing foundation for the subsequent construction of the ground elevation model.

[0055] (3) By fusing the elevation data set Ds after elimination, calculating the elevation fusion evaluation index Hf, and integrating to generate the ground elevation data set DEMb, the efficient fusion and spatial matching of multi-period data are completed, and the availability of the ground elevation data set DEMb is accurately judged. Through the trigger mechanism of the available state result, the data fusion quality is iteratively optimized. It can not only efficiently quantify the accuracy state of the ground elevation data, but also intelligently adjust the unavailable data through the iterative optimization mechanism, realizing the closed-loop management of the ground elevation data from quality assessment to model construction. Compared with traditional single error assessment methods, by adjusting multiple key thresholds, the accuracy and stability of data fusion are continuously optimized to ensure that the finally constructed ground elevation model has high precision and strong robustness under different environmental conditions, providing reliable support for large-scale and multi-period terrain mapping. Brief Description of the Drawings

[0056] Figure 1 Schematic diagram of the steps of a method for generating a ground elevation model based on co - orbiting satellites according to the present invention. Specific implementation manners

[0057] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0058] Embodiment 1

[0059] The present invention provides a method for generating a ground elevation model based on co - orbiting satellites. Please refer to Figure 1 , which includes the following steps:

[0060] S1. Collect multi - temporal elevation data of co - orbiting satellites, obtain relevant observation data, and perform pre - processing to form an elevation data set D;

[0061] S2. Extract features from the elevation data set D, then perform temporal consistency detection to judge the stability of the elevation data set D. By quantifying the consistency of multi - temporal data, obtain a consistency factor C, and perform correction according to the consistency factor C to obtain a corrected elevation data set Dc;

[0062] S3. Analyze short - term surface changes of the corrected elevation data set Dc, calculate a short - term elevation change index △H to dynamically eliminate interference points, and mark the interference points to obtain an eliminated elevation data set Ds after dynamic elimination;

[0063] S4. By fusing the eliminated elevation data sets Ds of different time series, obtain an elevation fusion evaluation index Hf, and perform integration to form a ground elevation data set DEMb, and judge the comparison with a preset precision allowable error threshold Tthe to obtain the available state result of the ground elevation data set DEMb;

[0064] S5. Perform iterative optimization and construct a ground elevation model according to the available state result of the ground elevation data set DEMb.

[0065] In this embodiment, elevation data for multiple time periods is collected and preprocessed to form an elevation data set D; based on the consistency factor C, temporal consistency detection and correction are performed on the elevation data set D to obtain a corrected elevation data set Dc; the short-term elevation change index ΔH is calculated to mark and eliminate dynamic interference points, and the elevation data set Ds after elimination is obtained; based on the calculation of the elevation fusion evaluation index Hf, multi-time period data is integrated to form a ground elevation data set DEMb, and by comparing it with a preset precision allowable error threshold Tthe, the available state result of the ground elevation data set DEMb is obtained; finally, through iterative optimization, the construction of a high-precision ground elevation model is completed, realizing the efficient fusion of multi-time period elevation data after dynamically eliminating interference points, and generating a stable, highly accurate ground elevation model that can reflect dynamic change characteristics. Compared with traditional methods, this scheme not only improves the accuracy and consistency of the ground elevation model, but also has strong robustness and adaptability, providing a scientific and efficient solution path for high-precision terrain modeling under complex terrain conditions.

[0066] Embodiment 2

[0067] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 , specifically: S1 includes S11 and S12;

[0068] S11. Collect multi-time period elevation data of the same-track satellite, record the elevation data Hij of each observation and related observation conditions, including the sampling time Ti and the illumination condition Li, and form an initial elevation data set Dr = {Hij, Ti, Li};

[0069] Among them, the elevation data Hij specifically represents the elevation value of the jth ground sampling point of the ith observation; the sampling time T specifically represents the observation timestamp of the ith sampling, used to identify the time sequence of the observation; the illumination condition Li represents the ambient illumination condition of the ith sampling.

[0070] S12. Preprocess the initial elevation data set Dr, including noise elimination, outlier correction, and data standardization, to generate an elevation data set D = {Hij(std), Ti, Li(std)}, where Hij(std) and Li(std) respectively represent the standardized elevation data and standardized illumination condition after data standardization preprocessing;

[0071] Among them, noise elimination is performed by using the upper and lower quartile method to detect and eliminate noise points in the initial elevation data set Dr; outlier correction is performed by using the linear interpolation method to correct outliers in the initial elevation data set Dr after noise elimination; data standardization is performed by using the min-max normalization method to normalize the initial elevation data set Dr after outlier correction to the range of 0 to 1.

[0072] In this embodiment, through the acquisition process, an initial elevation data set Dr is formed, providing comprehensive basic information support for the accuracy and multi-temporal fusion of subsequent data. Further, through noise removal, outlier correction, and data standardization processing, noise points in the data are effectively removed, abnormal data is corrected, and an elevation data set D is generated, significantly improving the integrity and quality of the elevation data; ensuring the consistency of multi-temporal data in subsequent fusion calculations. Compared with traditional methods, the present invention can more accurately remove observation noise and correct outliers, and at the same time, by comprehensively recording multi-dimensional observation parameters, enhance the environmental adaptability of elevation data, laying a solid foundation for the construction of a high-precision ground elevation model.

[0073] Embodiment 3

[0074] This embodiment is an explanatory description carried out in Embodiment 2. Please refer to Figure 1 , specifically: S2 includes S21 and S22;

[0075] S21. Extract features from the elevation data set D to obtain standardized elevation data Hij(std) and sampling time Ti, perform temporal consistency detection on multi-temporal data of the same ground sampling point j, and form a time series feature set Hj = {H1j(std), H2j(std), H3j(std)……, Hnj(std)} to judge the stability of the elevation data set D. By quantifying the consistency of multi-temporal data, a consistency factor C is obtained;

[0076] Among them, n represents the total number of sampling observations, and Hj represents the elevation time series feature of the ground sampling point j;

[0077] The consistency factor C is obtained through the following calculation formula:

[0078]

[0079] In the formula, Cj represents the consistency factor at the position of the ground sampling point j, σ(Hj) represents the standard deviation of the time series feature set Hj, specifically used to measure the degree of data fluctuation, and Hmax and Hmin respectively represent the upper limit elevation value and the lower limit elevation value at the position of the ground sampling point j;

[0080] Among them, the standard deviation σ(Hj) is obtained through the calculation formula. In the formula, represents the mean value of the standardized elevation data Hij(std) in the time series feature set.

[0081] S22. Compare the obtained consistency factor C with the preset consistency status threshold Cthe to obtain the execution status result of dataset correction, and correct the elevation dataset D according to the execution status result of dataset correction to obtain the corrected elevation dataset Dc;

[0082] The execution status result of the dataset correction is obtained through the following comparison method:

[0083] When the consistency factor C < the consistency status threshold Cthe, the execution status result of dataset correction is obtained as executing correction, including correcting the standardized elevation data Hij(std) of the ground sampling point j to the mean value Hj, and adding the replaced data of the ground sampling point j to the corrected elevation dataset Dc;

[0084] When the consistency factor C ≥ the consistency status threshold Cthe, the execution status result of dataset correction is obtained as not executing correction, and directly adding the standardized elevation data Hij(std) of the ground sampling point j to the corrected elevation dataset Dc.

[0085] In this embodiment, by extracting features from the elevation dataset D, the time series feature set Hj is obtained, and the consistency factor C is calculated based on the fluctuations of the time series data of the ground sampling point j. Through the quantification of the consistency factor C, the stability of the time series feature set is accurately evaluated, providing a scientific basis for subsequent correction processing. According to the comparison result of the consistency factor C and the preset consistency status threshold Cthe, the execution status result of dataset correction is generated, solving the problem of judgment errors caused by differences in data fluctuation amplitudes at different sampling positions and times. At the same time, according to the strict comparison mechanism of the consistency factor C and the consistency status threshold Cthe, stable data and data to be corrected are effectively distinguished, and targeted correction of data is accurately achieved. Compared with the traditional one-size-fits-all correction method, only the data with large fluctuations is processed by mean correction, which not only improves the overall consistency of the elevation data but also maximally retains the spatio-temporal characteristics of the original data, thus providing high-quality basic data support for constructing a high-precision ground elevation model.

[0086] Embodiment 4

[0087] This embodiment is an explanatory description based on Embodiment 3. Please refer to Figure 1 , specifically: The S3 includes S31 and S32;

[0088] S31. Analyze the short-term surface changes of the corrected elevation dataset Dc, including calculating the short-term fluctuations between adjacent time points of the same ground sampling point j, obtaining the short-term elevation change index △H, and dynamically removing interference points;

[0089] The short-term elevation change index △H is obtained through the following calculation formula:

[0090]

[0091] In the formula, △Hij represents the short-term elevation change index of the j-th position of the ground sampling point in the i-th observation. Hij(std) represents the standardized elevation data of the j-th position of the ground sampling point in the i-th observation in the corrected elevation dataset Dc. H(i + 1)j(std) represents the standardized elevation data of the j-th position of the ground sampling point in the (i + 1)-th observation. Ti and Tk represent adjacent sampling times, specifically representing the sampling time Ti and the sampling time T(i + 1).

[0092] S32. Compare the short-term elevation change index △Hij with the preset short-term change evaluation threshold Hthe to determine the interference state result Mij of the j-th position of the ground sampling point, and mark the interference point for the j-th position of the ground sampling point according to the interference state result Mij, and obtain the removed elevation data set Ds after dynamically removing the corrected elevation dataset Dc;

[0093] The interference state result Mij is obtained through the following comparison method:

[0094]

[0095] Among them, when the short-term elevation change index △Hij > the short-term change evaluation threshold Hthe, the interference state result Mij = 1, specifically indicating that the j-th position of the ground sampling point needs to be marked as an interference point;

[0096] When the short-term elevation change index △Hij ≤ the short-term change evaluation threshold Hthe, the interference state result Mij = 0, specifically indicating that the j-th position of the ground sampling point does not need to be marked as an interference point.

[0097] In this embodiment, the short-term dynamic change analysis of the corrected elevation data set Dc and the marking of interference points are realized, and the problem that short-term surface changes may introduce dynamic interference and reduce the accuracy of elevation data is solved. By calculating the short-term elevation change index ΔH, the short-term elevation fluctuation degree of the ground sampling point j between adjacent time points is quantified, and the abnormal changes caused by temporary factors such as vegetation cover and rainwater waterlogging within a short period of time are effectively captured. Based on the comparison between the short-term elevation change index ΔH and the preset short-term change evaluation threshold Hthe, the interference state result is generated. By marking and dynamically removing the interference points, the removed elevation data set Ds is generated, so as to eliminate the influence of dynamic interference points on the elevation data to the greatest extent, accurately capture the elevation fluctuation of the ground sampling point between adjacent time points, and realize the accurate dynamic removal based on the short-term change evaluation threshold Hthe. Compared with the traditional static removal method, it can dynamically mark and remove short-term abnormal points while retaining stable terrain change information, providing strong dynamic interference robustness support for elevation data. This method significantly improves the temporal consistency and data quality of the removed elevation data set Ds, laying a more reliable dynamic change processing foundation for the subsequent construction of the ground elevation model.

[0098] Embodiment 5

[0099] This embodiment is an explanatory description carried out in Embodiment 4, please refer to Figure 1 , specifically: S4 includes S41 and S42;

[0100] S41. By fusing the removed elevation data sets Ds of different time series, the elevation fusion evaluation index Hf is obtained and integrated to form the ground elevation data set DEMb = {Hf, j, Xj, Yj}, where j represents the ground sampling point, and Xj and Yj represent the horizontal and vertical coordinates of the ground sampling point;

[0101] The elevation fusion evaluation index Hf is obtained through the following calculation formula:

[0102]

[0103] In the formula, Hij(std) represents the standardized elevation data of the i-th observation of the ground sampling point j in the removed elevation data set Ds, and n represents the total number of sampling observations.

[0104] S42. By extracting the comparison state with the preset accuracy allowable error threshold Tthe from the ground elevation data set DEMb, the available state result of the ground elevation data set DEMb is obtained;

[0105] The available state result of the ground elevation data set DEMb is obtained through the following comparison method:

[0106]

[0107] Among them, represents an existential logical symbol, specifically indicating that there is at least one elevation fusion evaluation index Hf ≥ the allowable error threshold Tthe in the ground elevation dataset DEmb, and returns 1; represents a universal logical symbol, specifically indicating that all elevation fusion evaluation indices Hf in the ground elevation dataset DEmb satisfy < the allowable error threshold Tthe, and returns 0;

[0108] When Mdem = 1, the available status result of the ground elevation dataset DEmb is obtained as available;

[0109] When Mdem = 0, the available status result of the ground elevation dataset DEmb is obtained as unavailable.

[0110] The said S5 includes S51;

[0111] S51. Trigger and execute the iterative optimization mechanism according to the available status result of the ground elevation dataset DEmb. When the available status result of the ground elevation dataset DEmb is available, do not trigger and execute the iterative optimization mechanism, and give a prompt to construct the ground elevation model. When the available status result of the ground elevation dataset DEmb is unavailable, trigger and execute the iterative optimization mechanism, including adjusting the allowable error threshold Tthe, the short-term change evaluation threshold Hthe, and the consistency status threshold Cthe.

[0112] In this embodiment, by fusing and eliminating the elevation data set Ds, calculating the elevation fusion evaluation index Hf, and integrating and generating the ground elevation dataset DEmb, the efficient fusion and spatial matching of multi-temporal data are completed. Further, by comparing the logical state of the elevation fusion evaluation index Hf with the preset allowable error threshold Tthe, the availability of the ground elevation dataset DEmb is accurately judged. Through the trigger mechanism of the available status result, automatically adjust the allowable error threshold Tthe, the short-term change evaluation threshold Hthe, and the consistency status threshold Cthe for the unavailable ground elevation dataset, and iteratively optimize the data fusion quality. It can not only efficiently quantify the accuracy status of the ground elevation data, but also intelligently adjust the unavailable data through the iterative optimization mechanism, realizing the closed-loop management from quality assessment to model construction of the ground elevation data. Compared with the traditional single error assessment method, by adjusting multiple key thresholds, continuously optimize the accuracy and stability of data fusion, ensure that the finally constructed ground elevation model has high accuracy and strong robustness under different environmental conditions, and provide reliable support for large-scale and multi-temporal topographic surveying.

[0113] Although embodiments of the present invention have been shown and described, those of ordinary skill in the art will appreciate that various changes, modifications, substitutions and permutations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for generating a ground elevation model based on co-orbital satellites, characterized in that: The following steps are involved: S1, collect multi-period elevation data of satellites in the same orbit, obtain relevant observation data, and pre-process them to form an elevation data set D; S2, extract features from the elevation dataset D, and then perform time series consistency detection to determine the stability of the elevation dataset D. By quantifying the consistency of multi-period data, obtain the consistency factor C, and make corrections based on the consistency factor C to obtain the corrected elevation dataset Dc; S3, analyzing the short-term surface changes of the corrected elevation data set Dc, calculating the short-term elevation change index △H to dynamically remove interference points, and marking the interference points to obtain the removed elevation data set Ds after dynamic removal; The S3 includes S31 and S32; S31, analyzing the short-term surface changes of the corrected elevation data set Dc, including calculating the short-term fluctuations of the same ground sampling point j between adjacent time points, obtaining the short-term elevation change index △H, and dynamically removing interference points; The short-term elevation change index △H is obtained by the following calculation formula: Where △Hij represents the short-term elevation change index of the i-th observation ground sampling point j, Hij(std) represents the standardized elevation data of the i-th observation ground sampling point j in the corrected elevation data set Dc, H(i+1)j(std) represents the standardized elevation data of the (i+1)-th observation ground sampling point j, Ti and Tk represent adjacent sampling times, specifically representing the sampling time Ti and the sampling time T(i+1); S4, by fusing the removed elevation data sets Ds of different time series, obtaining the elevation fusion evaluation index Hf, and integrating them to form the ground elevation dataset DEMb, and comparing them with the preset accuracy allowable error threshold Tthe, obtaining the available status result of the ground elevation dataset DEMb; S5. Iteratively optimize and construct a ground elevation model based on the available status results of the ground elevation dataset DEMb.

2. The method for generating a ground elevation model based on co-orbital satellites according to claim 1, characterized in that: Said S1 includes S11 and S12; S11, collect multi-period elevation data of the same-orbit satellite, record the elevation data Hij of each observation and related observation conditions, including sampling time Ti and illumination conditions Li, to form an initial elevation data set Dr = {Hij, Ti, Li}; Among them, the elevation data Hij is specifically represented by the elevation value of the jth ground sampling point of the i-th observation; the sampling time Ti specifically represents the observation timestamp of the i-th sampling, which is used to identify the time sequence of the observation; the lighting condition Li represents the ambient lighting condition of the i-th sampling.

3. The method for generating a ground elevation model based on co-orbital satellites according to claim 2, characterized in that: S12, preprocessing the initial elevation data set Dr, including noise removal, outlier correction and data standardization, to generate an elevation data set D = {Hij(std), Ti, Li(std)}, where Hij(std) and Li(std) represent the standardized elevation data and standardized illumination conditions after data standardization preprocessing, respectively; Among them, noise removal is carried out by using the upper and lower quartile method to detect the initial elevation data set Dr and remove the noise points in the observation data; outlier correction is carried out by using the linear interpolation method to correct the outliers of the initial elevation data set Dr after noise removal; data standardization is carried out by using the minimum-maximum normalization method to normalize the initial elevation data set Dr after the outlier correction to the range of 0 to 1.

4. The method for generating a ground elevation model based on co-orbital satellites according to claim 1, characterized in that: The S2 includes S21 and S22; S21, extract features from the elevation dataset D, obtain standardized elevation data Hij(std) and sampling time Ti, perform time series consistency detection on multi-period data of the same ground sampling point j, and form a time series feature set Hj={H1j(std), H2j(std), H3j(std)..., Hnj(std)} to judge the stability of the elevation dataset D, and obtain the consistency factor C by quantifying the consistency of multi-period data; Where n represents the total number of sampling observations, Hj represents the elevation time series characteristics of ground sampling point j; The consistency factor C is obtained by the following calculation formula: In the formula, Cj represents the consistency factor of the ground sampling point j, σ(Hj) represents the standard deviation of the time series feature set Hj, which is specifically used to measure the degree of data fluctuation, Hmax and Hmin represent the upper and lower elevation values ​​at the ground sampling point j, respectively; Among them, the standard deviation σ(Hj) is obtained by The calculation formula is obtained, where: Represents the mean of the standardized elevation data Hij(std) in the time series feature set.

5. The method for generating a ground elevation model based on co-orbital satellites according to claim 4, characterized in that: S22, comparing the obtained consistency factor C with a preset consistency state threshold Cthe, obtaining a dataset correction execution state result, and correcting the elevation dataset D according to the dataset correction execution state result to obtain a corrected elevation dataset Dc; The dataset modification execution status result is obtained by the following comparison method: When the consistency factor C is less than the consistency state threshold Cthe, the result of the dataset correction execution state is obtained as execution correction, including correcting the standardized elevation data Hij(std) of the ground sampling point j to the mean value The data of the replaced ground sampling point j is added to the corrected elevation data set Dc; When the consistency factor C≥consistency state threshold Cthe, the dataset correction execution state result is obtained as no correction is performed, and the standardized elevation data Hij(std) of the ground sampling point j is directly added to the corrected elevation dataset Dc.

6. The method for generating a ground elevation model based on co-orbital satellites according to claim 1, characterized in that: S32, comparing the short-term elevation change index △Hij with the preset short-term change assessment threshold Hthe, determining the interference state result Mij of the ground sampling point j, and marking the interference point at the ground sampling point j according to the interference state result Mij, and obtaining the eliminated elevation data set Ds after the corrected elevation data set Dc is dynamically eliminated; The interference state result Mij is obtained by the following comparison method: Among them, when the short-term elevation change index △Hij> the short-term change assessment threshold Hthe, the interference state result Mij=1, which specifically indicates that the position of the ground sampling point j needs to be marked as an interference point; When the short-term elevation change index △Hij≤the short-term change assessment threshold Hthe, the interference state result Mij=0, which specifically indicates that the position of the ground sampling point j does not need to be marked as an interference point.

7. The method for generating a ground elevation model based on co-orbital satellites according to claim 1, characterized in that: The S4 includes S41 and S42; S41, by fusing the removed elevation data sets Ds of different time series, obtaining the elevation fusion evaluation index Hf, and integrating them to form a ground elevation data set DEMb = {Hf, j, Xj, Yj}, where j represents a ground sampling point, and Xj and Yj represent the horizontal axis coordinate and the vertical axis coordinate of the ground sampling point; The elevation fusion evaluation index Hf is obtained by the following calculation formula: Where Hij(std) represents the standardized elevation data of the jth ground sampling point after eliminating the i-th observation in the elevation data set Ds, and n represents the total number of sampling observations.

8. The method for generating a ground elevation model based on co-orbital satellites according to claim 7, characterized in that: S42, obtaining the available status result of the ground elevation dataset DEMb by extracting and comparing the status with the preset accuracy allowable error threshold Tthe from the ground elevation dataset DEMb; The available status results of the ground elevation dataset DEMb are obtained by the following comparison method: in, Indicates the presence of logical symbols, Represents all logical symbols; When Mdem=1, the available status of the ground elevation dataset DEMb is obtained as available; When Mdem=0, the available status of the ground elevation dataset DEMb is obtained as unavailable.

9. The method for generating a ground elevation model based on co-orbital satellites according to claim 1, characterized in that: The S5 includes S51; S51. Trigger the execution of an iterative optimization mechanism according to the available status result of the ground elevation dataset DEMb. When the available status result of the ground elevation dataset DEMb is available, the execution of the iterative optimization mechanism is not triggered, and a prompt is given to build a ground elevation model. When the available status result of the ground elevation dataset DEMb is unavailable, the execution of the iterative optimization mechanism is triggered, including adjusting the accuracy allowable error threshold Tthe, the short-term change assessment threshold Hthe and the consistency status threshold Cthe.

Images