A method and system for correcting elevation data errors of EGM2008 model

Abstract

The present application relates to the technical field of elevation data error correction, and specifically to an elevation data error correction method and system of the EGM2008 model, the method comprising: determining the elevation anomaly difference of each measuring point; determining the elevation anomaly error of each measuring point; determining the local contour distribution value of each measuring point; using a neural network model and based on the elevation anomaly error, local contour distribution value and elevation anomaly of all standard measuring points, in combination with the EGM2008 model, correcting the normal height of the error measuring point. The present application analyzes the topographic features and constructs the local contour distribution value of the measuring point, and corrects the elevation anomaly of the error measuring point in combination with the neural network model. The accuracy of the elevation anomaly calculated by the EGM2008 model and the precision of the normal height are improved.

Description

Technical Field

[0001] The present application relates to the technical field of elevation data error correction, and in particular to an elevation data error correction method and system for an EGM2008 model. Background Art

[0002] In the field of engineering surveying and mapping, elevation data is an important basis for making topographic maps and terrain analysis, and is very important for various engineering surveying, construction and design fields. However, the accuracy of elevation data will directly affect the monitoring of engineering survey, design and construction, and low-precision elevation data cannot meet the requirements of elevation data in practical applications, so improving the accuracy of elevation data has important practical significance.

[0003] In the prior art, GNSS elevation measurement generally uses the GNSS elevation fitting method to solve elevation data. However, this solution method is greatly affected by the fitting accuracy, which easily leads to the problem of large errors and low accuracy in the solved elevation data. It can only meet the topographic map drawing in the early survey stage of the project, and cannot meet the requirements of elevation data in practical applications. At the same time, the prior art can calculate the elevation anomaly of each measuring point through the EGM2008 model, but due to the influence of external factors and the elevation fitting accuracy in the actual measurement process, the accuracy of measuring the elevation anomaly is reduced, making it impossible to obtain the accurate normal height of the measuring point in the area to be surveyed. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method and system for correcting the error of elevation data of the EGM2008 model. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for correcting errors in elevation data of an EGM2008 model, the method comprising the following steps:

[0006] Obtain the geodetic height, normal height and elevation anomaly of all measuring points in the area to be surveyed;

[0007] According to the differences among the geoid height, normal height and elevation anomaly of each measuring point, the elevation anomaly difference of each measuring point is determined;

[0008] The elevation difference confusion degree of each measuring point is determined according to the difference between the elevation anomaly difference degree of each measuring point and the elevation anomaly difference degrees of all measuring points in the neighborhood of the corresponding measuring point, and the degree of confusion of the elevation anomaly difference degrees of each measuring point and all measuring points in its neighborhood;

[0009] Obtain the abnormal score value of the elevation difference confusion of each measuring point, and combine the discrete degree of the elevation difference confusion of each measuring point and all measuring points in its neighborhood, as well as the elevation difference confusion of each measuring point, to determine the elevation abnormal error of each measuring point;

[0010] Determine the local contour distribution value of each measuring point based on the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point, combined with the elevation anomaly error;

[0011] The elevation anomaly difference of all measurement points in the area to be surveyed is used as the input of the threshold segmentation algorithm. The measurement points with elevation anomaly difference higher than the segmentation threshold are used as error measurement points, and the measurement points with elevation anomaly difference lower than the segmentation threshold are used as standard measurement points.

[0012] A neural network is trained based on the elevation anomaly errors, local contour distribution values ​​and elevation anomalies of all standard measuring points. The elevation anomaly correction value of the error measuring point is obtained using the trained neural network based on the elevation anomaly errors and local contour distribution values ​​of the error measuring point. The normal height of the error measuring point is corrected in combination with the EGM2008 model and the geodetic height of the error measuring point.

[0013] Preferably, the method for determining the elevation anomaly difference of each measuring point is:

[0014] Take the absolute value of the difference between the geoid height and the normal height of each measuring point and record it as the first difference of each measuring point;

[0015] The elevation anomaly difference of each measuring point is the difference between the first difference of each measuring point and the elevation anomaly.

[0016] Preferably, the process of determining the elevation difference confusion of each measuring point is as follows:

[0017] Analyze the cumulative sum of the differences between the elevation anomaly difference of each measuring point and the elevation anomaly difference of all measuring points in the neighborhood of the corresponding measuring point, and record it as the first sum value of each measuring point;

[0018] Analyze the information entropy of the elevation anomaly difference of each measuring point and all measuring points in its neighborhood;

[0019] The elevation difference confusion of each measuring point is the result of the fusion of the first sum value and information entropy of each measuring point.

[0020] Preferably, the method for determining the elevation anomaly error of each measuring point is:

[0021] Determine the elevation anomaly dispersion of each measuring point based on the degree of dispersion of the elevation difference chaos of each measuring point and all measuring points in its neighborhood, combined with the anomaly score value;

[0022] Based on the elevation difference confusion and the elevation anomaly discreteness, the elevation anomaly error of each measuring point is determined.

[0023] Preferably, the expression of the elevation anomaly discreteness of each measuring point is: , where is the elevation anomaly dispersion of the i-th measuring point; is the abnormal score value of the i-th measurement point; is the degree of discreteness of the elevation difference chaos of the i-th measuring point and all measuring points in its neighborhood.

[0024] Preferably, the elevation anomaly error of each measuring point is the product of the elevation difference confusion and the elevation anomaly discreteness of each measuring point.

[0025] Preferably, the method for determining the local contour distribution value of each measuring point is:

[0026] Determine the elevation difference distribution value of each measuring point according to the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point;

[0027] The local contour distribution value of each measuring point is the inverse of the product of the elevation anomaly error and the elevation difference distribution value of each measuring point.

[0028] Preferably, the expression of the elevation difference distribution value of each measuring point is: ; In the formula, Represents the elevation difference distribution value of the i-th measuring point; represents the geodetic height of the ith measuring point; represents the geodetic height of the gth measuring point in the neighborhood of the i-th measuring point; M represents the number of all measuring points in the neighborhood of the i-th measuring point.

[0029] Preferably, the correction of the normal height of the error measurement point by combining the EGM2008 model and the geodetic height of the error measurement point comprises:

[0030] The elevation anomaly errors and local contour distribution values ​​of all standard measurement points in the area to be surveyed are used as training samples of the neural network, the elevation anomalies of all standard measurement points are used as labels of the training samples, and the trained neural network model is used as a prediction model for elevation anomalies;

[0031] The elevation anomaly errors and local contour distribution values ​​of all error measurement points in the area to be surveyed are used as the input of the elevation anomaly prediction model to obtain the elevation anomaly correction values ​​of all error measurement points;

[0032] The expression of the normal height correction value of each error measurement point in the area to be surveyed is: ; In the formula, is the normal height correction value of the uth error measurement point in the area to be surveyed; is the geodetic height of the uth error measurement point in the area to be surveyed; It is the elevation anomaly correction value of the u-th error measurement point in the area to be surveyed.

[0033] In the second aspect, an embodiment of the present application also provides an elevation data error correction system for an EGM2008 model, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned methods for correcting the elevation data error of the EGM2008 model are implemented.

[0034] This application has at least the following beneficial effects:

[0035] The present application uses the EGM2008 model to analyze the topographic and geomorphic features, and determines the elevation difference confusion of each measuring point according to the difference between the elevation anomaly difference of each measuring point and the elevation anomaly difference of all measuring points in the neighborhood of the corresponding measuring point, as well as the degree of confusion of the elevation anomaly difference of each measuring point and all measuring points in its neighborhood, which has the beneficial effect of accurately correcting the normal height error; determines the elevation anomaly error of each measuring point according to the degree of discreteness of the elevation difference confusion of each measuring point and all measuring points in its neighborhood, as well as the elevation difference confusion and anomaly score value of each measuring point, which has the beneficial effect of more accurately predicting the influence of terrain factors on the measurement results and more accurately correcting the measurement results; determines the local contour distribution value of each measuring point according to the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point, combined with the elevation anomaly error, and corrects the elevation anomaly of the error measuring point in combination with the neural network model, and obtains the normal height correction value of the error measuring point, which has the beneficial effect of improving the accuracy of the elevation anomaly.

[0036] This application takes into account that the elevation data measurement process is affected by the terrain influence characteristics and local contour characteristics. The rapid changes in terrain fluctuations in the local area will cause local changes in the gravity field, which will reduce the accuracy of the gravity model and the geoid, thereby causing the EGM2008 gravity model to calculate the elevation anomaly with low accuracy. Therefore, by analyzing the terrain influence characteristics and local contour characteristics as the influencing factors of the elevation anomaly, and using the machine learning method to correct the elevation anomaly of the error measurement point, the problem of low accuracy of the elevation anomaly calculated by the EGM2008 gravity model when it is greatly affected by external interference is avoided, and then the normal height of the error measurement point is corrected, thereby improving the accuracy of the elevation anomaly calculated by the EGM2008 model and the accuracy of the normal height. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 A flowchart of a method for correcting elevation data errors of an EGM2008 model provided in one embodiment of the present application;

[0039] Figure 2 A flow chart of the steps for obtaining the degree of confusion of elevation difference provided in one embodiment of the present application;

[0040] Figure 3 A schematic diagram of a local contour distribution value extraction process provided for an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the method and system for correcting the error of elevation data of the EGM2008 model proposed by the present application, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0043] The following is a detailed description of a method and system for correcting elevation data errors of the EGM2008 model provided by the present application in conjunction with the accompanying drawings.

[0044] See also Figure 1 , which shows a flowchart of a method for correcting elevation data errors of an EGM2008 model provided by an embodiment of the present application, the method comprising the following steps:

[0045] Step S1: Use GNSS measurement technology and leveling joint measurement technology to obtain the normal height, geodetic height, and GNSS coordinate geodetic height of the measurement point.

[0046] The area to be surveyed is divided into n sub-areas, and the center position of each sub-area is taken as a measurement point.

[0047] It should be noted that the value of the sub-region n is artificially set. In this embodiment, the value of the sub-region n is 100. The implementer can also set it by himself according to the specific situation. This embodiment does not impose any special limitation.

[0048] The GNSS measurement technology is used to measure the geodetic height and GNSS plane coordinates of each measuring point in the surveying area to obtain the geodetic height and GNSS coordinates of each measuring point in the area to be detected.

[0049] At the same time, a level is used to measure the normal height of each measuring point in the surveying area in a leveling joint survey manner to obtain the normal height of each measuring point in the area to be detected.

[0050] Among them, GNSS measurement and leveling joint measurement are both well-known technologies, and the specific processes will not be repeated here.

[0051] Step S2: Determine the elevation anomaly difference of each measuring point according to the difference between the geoid height, normal height and elevation anomaly of each measuring point.

[0052] In the fields of engineering surveying, topographic mapping, etc., very accurate terrain elevation data is required. In order to more accurately determine the elevation of ground points, it is necessary to analyze the elevation anomaly of each measuring point in the area to be surveyed.

[0053] Elevation anomaly generally refers to the elevation difference between the geoid and the reference ellipsoid. The existing technology can calculate the elevation anomaly of each measuring point through the EGM2008 model.

[0054] The geodetic height and GNSS coordinates of each measuring point in the area to be surveyed are used as the input of the EGM2008 model, and the elevation anomaly of each measuring point in the area to be surveyed is calculated by the EGM2008 model.

[0055] Among them, the calculation of elevation anomaly by the EGM2008 model is a well-known technology, and the specific process will not be repeated here.

[0056] However, due to the influence of external factors and the accuracy of elevation anomaly in the actual measurement process, the accuracy of measuring elevation anomaly is likely to be low. Therefore, it is necessary to analyze the changes in elevation anomaly of the measuring points in the surveying area to improve the accuracy and reliability of subsequent error correction of elevation anomaly.

[0057] In order to analyze the accuracy of elevation anomaly, the elevation anomaly difference of each measuring point is determined according to the difference between the geoid height, normal height and elevation anomaly of each measuring point, specifically:

[0058] Take the absolute value of the difference between the geoid height and the normal height of each measuring point and record it as the first difference of each measuring point;

[0059] The greater the difference between the elevation anomaly discrimination value and the elevation anomaly of the measuring point, the smaller the accuracy of the elevation anomaly of the measuring point, and the greater the impact on the subsequent normal height error correction.

[0060] The elevation anomaly difference of each measuring point is the difference between the first difference of each measuring point and the elevation anomaly.

[0061] It should be noted that there are many methods for measuring differences. In this embodiment, the difference between the first difference and the elevation anomaly is measured by analyzing the difference between the first difference and the elevation anomaly. The implementer may also use other methods that can measure the difference, such as ratio, and this embodiment does not impose any special restrictions.

[0062] The greater the difference between the elevation anomaly discrimination value and the elevation anomaly, the more it indicates that the measurement process is affected by external factors or the accuracy of the elevation anomaly, resulting in a smaller accuracy of the elevation anomaly at the measuring point, and the greater the elevation anomaly difference; conversely, the smaller the difference between the elevation anomaly discrimination value and the elevation anomaly, the less it indicates that the measurement process is affected by external factors or the accuracy of the elevation anomaly, and the greater the accuracy of the elevation anomaly at the measuring point.

[0063] Step S3: Determine the elevation difference confusion degree of each measuring point according to the difference between the elevation anomaly difference degree of each measuring point and the elevation anomaly difference degrees of all measuring points in the neighborhood of the corresponding measuring point, and the degree of confusion of the elevation anomaly difference degrees of each measuring point and all measuring points in its neighborhood.

[0064] Generally speaking, the complexity of topography will affect the measurement of normal height. If the variation of elevation anomaly difference in a certain adjacent area is more chaotic, it will more easily affect the accuracy of elevation anomaly, thus resulting in lower accuracy of normal height. In order to accurately correct the error of normal height, it is necessary to analyze the impact of topography complexity on normal height measurement.

[0065] Therefore, based on the difference between the elevation anomaly difference of each measuring point and the elevation anomaly difference of all measuring points in the neighborhood of the corresponding measuring point, as well as the degree of confusion of the elevation anomaly difference of each measuring point and all measuring points in its neighborhood, the elevation difference confusion of each measuring point is determined, specifically:

[0066] A neighborhood W is divided with each measuring point as the center, wherein the value of the neighborhood W is artificially set. In this embodiment, the value of the neighborhood W is 8, and the implementer can also set it by himself according to the specific situation. This embodiment does not impose any special restrictions.

[0067] Analyze the cumulative sum of the differences between the elevation anomaly difference of each measuring point and the elevation anomaly difference of all measuring points in the neighborhood of the corresponding measuring point, and record it as the first sum value of each measuring point;

[0068] It should be noted that there are many methods for measuring differences. In this embodiment, the absolute value of the difference between the elevation anomaly difference of each measuring point and the elevation anomaly difference of its adjacent points is taken to measure the difference between the elevation anomaly differences of adjacent measuring points. The implementer may also adopt other methods of measuring differences such as ratios according to specific circumstances, and this embodiment does not impose any special restrictions.

[0069] Furthermore, the information entropy of the elevation anomaly difference of each measuring point and all measuring points in its neighborhood is analyzed;

[0070] The elevation difference confusion of each measuring point is the result of the fusion of the first sum value and information entropy of each measuring point.

[0071] The calculation process of information entropy is a well-known technology, and the specific calculation process will not be repeated here.

[0072] It should be understood that fusion refers to combining two or more indicators by adding or multiplying them together in order to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analysis methods. The implementer can choose according to the specific situation, and this embodiment does not impose any special restrictions.

[0073] Preferably, in the present embodiment, the elevation difference confusion degree of each measuring point is the product of the first sum of each measuring point and the information entropy; in actual application, as other implementation methods, the elevation difference confusion degree of each measuring point is an exponential function value with a natural constant as the base and the addition result of the first sum of each measuring point and the information entropy as the independent variable.

[0074] The greater the difference between the elevation anomaly difference of any measuring point and the elevation anomaly difference of any measuring point in its neighborhood, that is, the larger the first sum value, and the larger the information entropy of the corresponding measuring point, the greater the change in the elevation anomaly difference between the measuring point and the measuring points in its neighborhood, and the higher the confusion of the elevation anomaly difference between the measuring point and all the measuring points in its neighborhood, which to a certain extent reflects that the terrain and landform at the location of the measuring point are more complex, which is easy to affect the accuracy of the elevation anomaly, and the greater the elevation difference confusion; conversely, if the first sum value of any measuring point is smaller and the information entropy of the corresponding measuring point is smaller, it means that the terrain and landform at the location of the measuring point are relatively simple, and the impact on the accuracy of the elevation anomaly is smaller.

[0075] Preferably, the elevation difference confusion degree acquisition step flow chart provided in this embodiment is as follows: Figure 2 shown.

[0076] Step S4: Obtain the abnormal score value of each measuring point regarding the elevation difference confusion, and determine the elevation abnormal error of each measuring point by combining the discrete degree of the elevation difference confusion of each measuring point and all measuring points in its neighborhood, as well as the elevation difference confusion of each measuring point.

[0077] Normally, the complexity of the terrain and landforms at different locations in the same surveying area is highly similar. However, due to the undulating changes in the terrain and landforms, such as hilly areas, which are generally distributed in the transition zone between mountains and plains, showing undulating terrain features, the terrain features at the edge of the surveying area change significantly, and the similarity between the complexity of the terrain and landforms will be broken. In order to more accurately correct the error of the normal height, it is necessary to analyze the similar features between the terrain and landforms, specifically:

[0078] (1) In order to analyze the complexity of the topography in the local area where the measurement point is located, the elevation difference confusion of all measurement points in the area to be surveyed is used as the input of the anomaly scoring algorithm to obtain the anomaly score value of each measurement point to determine the similarity between the topography of the measurement point and other measurement points in the local area.

[0079] It should be noted that there are many commonly used anomaly detection algorithms. In this embodiment, the LOF local outlier factor algorithm is used to obtain the anomaly score value of each measuring point. The implementer can also use the z-score algorithm to obtain the anomaly score value of each measuring point regarding the elevation difference confusion. This embodiment does not impose any special restrictions.

[0080] Among them, the LOF local outlier factor algorithm is a well-known technology in the field of anomaly detection, and the specific process of performing anomaly detection on data and obtaining anomaly score values ​​will not be repeated here.

[0081] The larger the anomaly score, the smaller the similarity between the topography of the measuring point and that of other measuring points in its local range, the higher the degree of change in the topography of the measuring point, and the more likely it is to affect the accuracy of the elevation anomaly.

[0082] In order to more accurately understand the terrain characteristics of the area where each measuring point is located, the elevation anomaly dispersion of each measuring point is determined based on the discrete degree of the elevation difference confusion of each measuring point and all measuring points in its neighborhood, combined with the anomaly score value, to determine the degree of influence of the terrain characteristics on the measurement data, thereby improving the correction accuracy of the measured elevation, specifically:

[0083] The expression of the elevation anomaly dispersion of each measuring point is: , where is the elevation anomaly dispersion of the i-th measuring point; is the abnormal score value of the i-th measurement point; is the degree of discreteness of the elevation difference chaos of the ith measuring point and all measuring points in its neighborhood.

[0084] It should be noted that there are many methods to measure the degree of discreteness of a set of data. In this embodiment, the degree of discreteness of the elevation difference confusion is measured by calculating the dispersion coefficient of the elevation difference confusion of each measuring point and all measuring points in its neighborhood. The implementer can also use other methods that can measure the degree of discreteness, such as variance and standard deviation, and this embodiment does not impose any special restrictions.

[0085] The calculation process of the discrete coefficient is a well-known technique, and the specific calculation process will not be described in detail.

[0086] The larger the anomaly score value, the smaller the similarity between the topography of the measuring point and the topography of other measuring points in its local range. At the same time, the larger the dispersion coefficient, the greater the discrete change of the elevation difference chaos within the neighborhood of the measuring point, and the more likely the sub-region where the measuring point is located is to be in an area where the terrain features have changed significantly. At this time, the smaller the similarity between the topography and the topography, the greater the dispersion of the elevation anomaly; conversely, the smaller the anomaly score value, the greater the similarity between the topography of the measuring point and the topography of other measuring points in its local range. At the same time, the smaller the dispersion coefficient, the smaller the discrete change of the elevation difference chaos within the neighborhood of the measuring point, and the less likely the sub-region where the measuring point is located is to be in an area where the terrain features have changed significantly. At this time, the greater the similarity between the topography and the topography, the smaller the dispersion of the elevation anomaly.

[0087] (2) Furthermore, in order to more accurately predict the impact of terrain factors on the measurement results, the elevation anomaly error of each measurement point is determined based on the elevation difference confusion and the elevation anomaly discreteness, so as to make a more accurate correction to the measurement results, specifically:

[0088] The elevation anomaly error of each measuring point is the product of the elevation difference confusion and the elevation anomaly dispersion of each measuring point.

[0089] The higher the elevation difference confusion of the measuring point and the smaller the elevation anomaly discreteness, the larger the elevation anomaly error of the measuring point, which indicates that the terrain and geomorphic changes in the local surveying and mapping area of ​​the measuring point are more complex, and the more likely it is to affect the accuracy of the elevation anomaly, and the larger the elevation anomaly error; conversely, the lower the elevation difference confusion of the measuring point and the larger the elevation anomaly discreteness, the smaller the elevation anomaly error of the measuring point, and the elevation difference confusion of the measuring point indicates that the terrain and geomorphic changes in the local surveying and mapping area of ​​the measuring point are simpler, the influence of terrain and geomorphic changes on the accuracy of the elevation anomaly is smaller, and the elevation anomaly error is smaller.

[0090] Step S5: Determine the local contour distribution value of each measuring point based on the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point and the elevation anomaly error.

[0091] In general, the vicinity of the measuring point has a strong contour distribution feature, that is, the elevation of the measuring point in the local area has good consistency. When the change in the geoid height in the local area is greater, it is more likely to be affected by the fitting accuracy, causing larger elevation fluctuations within the local surveying range. At this time, the degree of error correction for the measuring point is greater.

[0092] Therefore, based on the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point, combined with the elevation anomaly error, the local contour distribution value of each measuring point is determined to correct the normal height of the measuring point, specifically:

[0093] (1) According to the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point, the elevation difference distribution value of each measuring point is determined, specifically:

[0094] The expression of the elevation difference distribution value of each measuring point is: ; In the formula, Represents the elevation difference distribution value of the i-th measuring point; represents the geodetic height of the ith measuring point; represents the geodetic height of the gth measuring point in the neighborhood of the i-th measuring point; M represents the number of all measuring points in the neighborhood of the i-th measuring point.

[0095] The greater the difference between the geoid heights within the local surveying and mapping range of the measuring point, to a certain extent, it indicates that the local area contour distribution of the measuring point has a larger singular change, and the smaller the local contour distribution characteristics, the larger the elevation difference distribution value; conversely, the smaller the difference between the geoid heights within the local surveying and mapping range of the measuring point, to a certain extent, it indicates that the local area contour distribution of the measuring point has a smaller singular change, and the larger the local contour distribution characteristics, the smaller the elevation difference distribution value.

[0096] (2) Generally speaking, the greater the terrain influence characteristics of the measuring point, the more likely it is to lead to poor accuracy of the elevation anomaly. At the same time, the smaller the local contour characteristics of the measuring point, to a certain extent, the lower the accuracy of the elevation anomaly, and the greater the degree of error correction for the measuring point.

[0097] Therefore, based on the elevation difference distribution value and elevation anomaly error of each measuring point, the local contour distribution value of each measuring point is determined, specifically:

[0098] The local contour distribution value of each measuring point is the inverse of the product of the elevation anomaly error and the elevation difference distribution value of each measuring point.

[0099] The greater the terrain influence characteristics of the measuring point and the greater the local contour characteristics, that is, the greater the elevation anomaly error and elevation difference distribution value of the measuring point, the smaller the contour distribution characteristics on the local terrain of the measuring point, the lower the accuracy of the elevation anomaly at the corresponding measuring point, and the smaller the local contour distribution value; conversely, the smaller the elevation anomaly error and elevation difference distribution value of the measuring point, the greater the local contour distribution value.

[0100] Preferably, the schematic diagram of the local contour distribution value extraction process provided in this embodiment is as follows: Figure 3 shown.

[0101] Step S6: Adopting the neural network model and based on the elevation anomaly errors, local contour distribution values ​​and elevation anomalies of all standard measurement points, combined with the EGM2008 model, the normal height of the error measurement point is corrected.

[0102] (1) In order to extract measurement points with higher elevation anomaly accuracy, the elevation anomaly differences of all measurement points in the area to be surveyed are used as the input of the threshold segmentation algorithm to obtain the segmentation threshold.

[0103] Furthermore, the measurement points corresponding to the elevation anomaly differences below the segmentation threshold are used as standard measurement points; and the measurement points corresponding to the elevation anomaly differences above the segmentation threshold are used as error measurement points.

[0104] It should be noted that there are many commonly used threshold segmentation algorithms. In this embodiment, the maximum inter-class variance algorithm is used to classify the measurement points in the surveying area. The implementer can also use a clustering-based threshold segmentation algorithm to classify the measurement points according to the specific situation. This embodiment does not impose any special restrictions.

[0105] Among them, the maximum inter-class variance algorithm is a well-known technology, and its segmentation principle will not be described in detail.

[0106] (2) Furthermore, in order to correct the elevation anomaly of the erroneous measurement point, the elevation anomaly errors of all standard measurement points in the area to be surveyed are combined into an elevation anomaly error sequence, the local contour distribution values ​​of all standard measurement points are combined into a local contour distribution sequence, and the elevation anomalies of all standard measurement points are combined into an elevation anomaly sequence.

[0107] The elevation anomaly error sequence and the local contour distribution sequence are used as training samples, and the elevation anomaly sequence is used as the label sequence of the training samples;

[0108] The training samples and their label sequences are used as the input of the long short-term memory neural network model, and the long short-term memory neural network model is trained. The root mean square loss function and gradient descent method are used as the loss function and optimization algorithm of the long short-term memory neural network model respectively, and the trained model is used as a prediction model for elevation anomaly.

[0109] Among them, the training process of the long short-term memory neural network model, the root mean square loss function and the gradient descent method are all well-known technologies, and their specific working principles will not be repeated here.

[0110] (3) Furthermore, the elevation anomaly errors and local contour distribution values ​​of all the error measurement points in the area to be surveyed are used as inputs of the elevation anomaly prediction model, and the prediction results are the elevation anomaly correction values ​​of the error measurement points in the area to be surveyed.

[0111] Among them, the process of using the elevation anomaly prediction model to make predictions is a well-known technology, and the specific process of the prediction will not be repeated here.

[0112] (4) In order to correct the normal height of the error measurement point in the area to be surveyed, the elevation anomaly correction value of each error measurement point in the area to be surveyed is used as the input of the EGM2008 model. The EGM2008 model is ; Among them, h is the normal height; H is the earth height; It is an elevation anomaly.

[0113] The EGM2008 model is used to calculate the normal height correction value of each error measurement point in the area to be surveyed, specifically:

[0114] The expression of the normal height correction value of each error measurement point in the area to be surveyed is: ; In the formula, is the normal height correction value of the uth error measurement point in the area to be surveyed; is the geodetic height of the uth error measurement point in the area to be surveyed; It is the elevation anomaly correction value of the u-th error measurement point in the area to be surveyed.

[0115] When the external interference factors are large, the difference between the elevation anomaly calculated by the EGM2008 model and its real elevation anomaly is large, which makes the accuracy of the subsequent measurement of the normal height poor. This embodiment corrects the elevation anomaly of the error measurement point that is prone to measurement error by analyzing the terrain influence characteristics of the measurement point and the local contour distribution characteristics, and obtains a more accurate elevation anomaly, that is, the elevation anomaly correction value; the EGM2008 model is used and based on the elevation anomaly correction value, the normal height correction value of the error measurement point is calculated, that is, the error correction of the normal height of the error measurement point is performed, and the error correction of the elevation data is completed, thereby improving the accuracy of the elevation data.

[0116] Based on the same inventive concept as the above method, an embodiment of the present application also provides an elevation data error correction system for an EGM2008 model, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods for correcting elevation data errors of the EGM2008 model when executing the computer program.

[0117] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0118] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0119] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for correcting the elevation data error of the EGM2008 model, characterized in that: The method comprises the following steps: Obtain the geodetic height, normal height and elevation anomaly of all measuring points in the area to be surveyed; According to the differences among the geoid height, normal height and elevation anomaly of each measuring point, the elevation anomaly difference of each measuring point is determined; The elevation difference confusion degree of each measuring point is determined according to the difference between the elevation anomaly difference degree of each measuring point and the elevation anomaly difference degrees of all measuring points in the neighborhood of the corresponding measuring point, and the degree of confusion of the elevation anomaly difference degrees of each measuring point and all measuring points in its neighborhood; Obtain the abnormal score value of the elevation difference confusion of each measuring point, and combine the discrete degree of the elevation difference confusion of each measuring point and all measuring points in its neighborhood, as well as the elevation difference confusion of each measuring point, to determine the elevation abnormal error of each measuring point; Determine the local contour distribution value of each measuring point based on the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point, combined with the elevation anomaly error; The elevation anomaly difference of all measurement points in the area to be surveyed is used as the input of the threshold segmentation algorithm. The measurement points with elevation anomaly difference higher than the segmentation threshold are used as error measurement points, and the measurement points with elevation anomaly difference lower than the segmentation threshold are used as standard measurement points. Based on the elevation anomaly errors, local contour distribution values ​​and elevation anomalies of all standard measurement points, a neural network is trained. The elevation anomaly correction value of the error measurement point is obtained by using the trained neural network and based on the elevation anomaly errors and local contour distribution values ​​of the error measurement point. The normal height of the error measurement point is corrected in combination with the EGM2008 model and the geodetic height of the error measurement point. The determination process of the elevation difference confusion degree of each measuring point is as follows: Analyze the cumulative sum of the differences between the elevation anomaly difference of each measuring point and the elevation anomaly difference of all measuring points in the neighborhood of the corresponding measuring point, and record it as the first sum value of each measuring point; Analyze the information entropy of the elevation anomaly difference of each measuring point and all measuring points in its neighborhood; The elevation difference confusion of each measuring point is the result of the fusion of the first sum value and information entropy of each measuring point; The method for determining the elevation anomaly error of each measuring point is: Determine the elevation anomaly dispersion of each measuring point based on the degree of dispersion of the elevation difference chaos of each measuring point and all measuring points in its neighborhood, combined with the anomaly score value; Determining the elevation anomaly error of each measuring point based on the elevation difference confusion and the elevation anomaly dispersion; The expression of the elevation anomaly discreteness of each measuring point is: , where is the elevation anomaly dispersion of the i-th measuring point; is the abnormal score value of the i-th measurement point; is the degree of discreteness of the elevation difference chaos of the i-th measuring point and all measuring points in its neighborhood.

2. The method for correcting the elevation data error of the EGM2008 model according to claim 1, characterized in that: The method for determining the elevation anomaly difference of each measuring point is: Take the absolute value of the difference between the geoid height and the normal height of each measuring point and record it as the first difference of each measuring point; The elevation anomaly difference of each measuring point is the difference between the first difference of each measuring point and the elevation anomaly.

3. The method for correcting the elevation data error of the EGM2008 model according to claim 1, characterized in that: The elevation anomaly error of each measuring point is the product of the elevation difference confusion and the elevation anomaly dispersion of each measuring point.

4. The method for correcting elevation data errors of the EGM2008 model according to claim 1, characterized in that: The method for determining the local contour distribution value of each measuring point is as follows: Determine the elevation difference distribution value of each measuring point according to the average distribution of the difference between the geodetic height of each measuring point and the geodetic height of all measuring points in the neighborhood of the corresponding measuring point; The local contour distribution value of each measuring point is the inverse of the product of the elevation anomaly error and the elevation difference distribution value of each measuring point.

5. The method for correcting the elevation data error of the EGM2008 model as claimed in claim 4, characterized in that: The expression of the elevation difference distribution value of each measuring point is: ; In the formula, Represents the elevation difference distribution value of the i-th measuring point; represents the geodetic height of the ith measuring point; represents the geodetic height of the gth measuring point in the neighborhood of the i-th measuring point; M represents the number of all measuring points in the neighborhood of the i-th measuring point.

6. The method for correcting the elevation data error of the EGM2008 model as claimed in claim 1, characterized in that: The correction of the normal height of the error measurement point by combining the EGM2008 model and the geodetic height of the error measurement point includes: The elevation anomaly errors and local contour distribution values ​​of all standard measurement points in the area to be surveyed are used as training samples of the neural network, the elevation anomalies of all standard measurement points are used as labels of the training samples, and the trained neural network model is used as a prediction model for elevation anomalies; The elevation anomaly errors and local contour distribution values ​​of all error measurement points in the area to be surveyed are used as the input of the elevation anomaly prediction model to obtain the elevation anomaly correction values ​​of all error measurement points; The expression of the normal height correction value of each error measurement point in the area to be surveyed is: ; In the formula, is the normal height correction value of the uth error measurement point in the area to be surveyed; is the geodetic height of the uth error measurement point in the area to be surveyed; It is the elevation anomaly correction value of the u-th error measurement point in the area to be surveyed.

7. An elevation data error correction system for an EGM2008 model, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the elevation data error correction method of the EGM2008 model as described in any one of claims 1-6 are implemented.

Images