Airborne gravity data leveling method without rejecting gross errors
By employing median filtering with a moving window and polynomial fitting in airborne gravity data, the problems of low efficiency and reliance on experience caused by gross errors in airborne gravity measurements were solved, achieving efficient data leveling without the need to remove gross errors.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- AIRBORNE SURVEY & REMOTE SENSING CENTER OF NUCLEAR IND
- Filing Date
- 2023-10-26
- Publication Date
- 2026-07-14
Smart Images
Figure CN117631066B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to an airborne gravity data processing technology, specifically an airborne gravity data leveling method that does not require the removal of gross errors. Background Technology
[0002] Airborne gravity measurement is a method that uses an aircraft as a carrier to measure near-Earth gravitational acceleration by comprehensively applying gravimeters, inertial navigation systems, GPS, and altimeters and attitude measuring devices. It can quickly and economically acquire high-frequency local anomaly information in a uniformly distributed, accurate, and large-area gravity field, which can be used to study the Earth's internal structure and material composition, and to solve problems in geological prospecting and environmental studies. However, airborne gravity measurements are affected by many interference factors, resulting in differences in horizontal baseline values in the original airborne free-space gravity anomaly data between different measurement lines. Data leveling is required before interpretation.
[0003] The commonly used leveling method is the tangent-line leveling method, which uses the measurement difference (intersection discrepancy) at the intersection of the measuring line and the tangent line to adjust the level between the measuring lines. When using the tangent-line method for leveling, the calculated intersection discrepancy often contains gross errors caused by measurement mistakes (i.e., the absolute value of the intersection discrepancy is large). The presence of gross errors will severely distort the fitted curve of the intersection discrepancy, affecting the data processing results.
[0004] Therefore, gross errors need to be removed before leveling the cutting lines. However, the amount of airborne gravity measurement data is large, and there are many gross error points to be removed. If the gross error points are removed line by line, it will take a lot of time. At the same time, the removal of gross errors also depends on the experience of the data processor. Summary of the Invention
[0005] The purpose of this invention is to provide a method for leveling airborne gravity data without removing gross errors, so as to solve the problems of large workload, low efficiency and reliance on the experience of data processors in the current leveling method using the cutting line method.
[0006] The present invention is implemented as follows: a method for leveling airborne gravity data without removing gross errors, including data measurement and preparation, leveling of airborne free-space gravity anomaly data along the cutting line, and leveling of airborne free-space gravity anomaly data along the main survey line.
[0007] S1, data measurement and preparation, includes the following steps:
[0008] S11, Lay out the main survey line and the cutting line perpendicular to the direction of the main survey line in the work area;
[0009] S12, conduct airborne gravity measurements and collect data according to the established survey lines;
[0010] S13 processes the raw airborne gravity data collected on the main survey line and the cutting line to obtain airborne free space gravity anomaly data.
[0011] S2, leveling of free-space gravity anomaly data for cutting lines includes the following steps:
[0012] S21, search for all intersections between the cutting line and the main survey line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomaly values between the cutting line and the main survey line at the intersections, and calculate the discrepancy value of the intersections on the cutting line;
[0013] S22, on the cutting line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points. The result is used as the optimal compensation value for the airborne free space gravity anomaly at each measuring point on the cutting line and is used for compensation.
[0014] S3, Leveling of airborne free-space gravity anomaly data along the main survey line, including the following steps:
[0015] S31, search for all intersections of the main survey line and the cutting line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomaly values of the main survey line and the leveled cutting line at the intersections, and calculate the discrepancy value of the intersections on the main survey line.
[0016] S32. On the main measurement line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points. The result is used as the optimal compensation value for the airborne free space gravity anomaly at the corresponding measurement points on the main measurement line and compensation is performed.
[0017] It also includes step S4, leveling result verification: calculate the vertical first derivative of the leveled free-space gravity anomaly data and plot the leveling result.
[0018] In step S21, the position of the intersection point between the cutting line and the main measuring line is determined. Using the cutting line as a reference, the measuring point on the cutting line that is closest to the intersection point is searched and the measuring point is taken as the intersection point. The field value of the cutting line at the intersection point remains unchanged, and the field value of the main measuring line at the intersection point is obtained by linear interpolation.
[0019] In step S31, the position of the intersection point between the main test line and the leveled cutting line is determined. Using the main test line as a reference, the test point closest to the intersection point on the main test line is searched and the test point is taken as the intersection point. The field value of the main test line at the intersection point remains unchanged, and the field value of the cutting line at the intersection point is obtained by linear interpolation.
[0020] In steps S21 and S31, the measuring points on the cutting line and the main measuring line are fitted using a polynomial fitting method to determine the approximate intersection point of the main measuring line and the cutting line or the approximate intersection point of the cutting line and the main measuring line.
[0021] The median filtering method based on the moving window is used to filter the discrepancies at intersection points. The calculation formula is as follows:
[0022] yi=Median xi =(xi-v,…xi…, xi+v), i∈Z, v=(n-1) / 2
[0023] Where n is the size of the moving window, taking an odd value; xi is the discrepancy value of the i-th intersection point, and yi is the discrepancy value of the i-th intersection point after median filtering based on the moving window.
[0024] The moving window size n must cover at least the data segment containing three consecutive intersections.
[0025] Processing of raw airborne gravity data includes carrier motion acceleration correction, Etherworth correction, eccentricity correction, altitude correction, normal gravity value correction, zero-point drift correction, and low-pass filtering.
[0026] The distance between cutting lines should be 5 to 10 times the spacing between main test lines, and the test line length should be extended by at least one filter length.
[0027] This invention employs a static cutting line leveling method and treats gross errors in the array of crosspoint discrepancies as random interference. It performs median filtering based on a moving window on the crosspoint discrepancies, thereby leveling the measured airborne gravity data. During the leveling process, there is no need to remove data containing gross errors, saving data processing time and solving the problem of dependence on the processor's experience during the leveling process. Attached Figure Description
[0028] Figure 1 This is a flowchart of the present invention.
[0029] Figure 2 This is a diagram showing the arrangement of airborne geophysical survey lines according to the present invention.
[0030] Figure 3 (a) is the airborne gravity measurement data of the present invention. Figure 3 (b) is the processed free-space gravity anomaly data obtained after processing.
[0031] Figure 4 It is a distribution map of crosspoint discrepancies containing gross errors.
[0032] Figure 5 This is a schematic diagram of the median filtering principle of the present invention.
[0033] Figure 6 This is a schematic diagram of the density distribution of measuring points and intersection points.
[0034] Figure 7 This is a graph showing the median filtering result of the discrepancies at the intersection points on a certain measurement line, and the result after smoothing by low-pass filtering.
[0035] Figure 8 This is a gravity anomaly diagram of free space before leveling.
[0036] Figure 9 This is a gravity anomaly map of free space after being leveled using median filtering.
[0037] Figure 10 This is a plot of the vertical first derivative of the gravity anomaly in free space after leveling. Detailed Implementation
[0038] like Figure 1 As shown, the present invention is a method for leveling airborne gravity data without removing gross errors, including steps such as data measurement and preparation, leveling of airborne free space gravity anomaly data along the cutting line, and leveling of airborne free space gravity anomaly data along the main survey line.
[0039] The first step is to measure and prepare the data.
[0040] The main survey line and the cutting line perpendicular to the main survey line are laid out in the work area.
[0041] The main survey lines and cutting lines are laid out according to the geological tasks and the size of the objects to be explored in the survey area. The spacing between the cutting lines is determined according to the task requirements and the characteristics of the work area. The spacing between the cutting lines is 5 to 10 times that of the spacing between the main survey lines, and the length of the survey lines is extended by at least one filter length.
[0042] Specifically, such as Figure 2 As shown, the spacing between the main survey lines and the tangent lines in this invention is 2km and 10km respectively. To avoid edge effects caused during the low-pass filtering of airborne gravity data, both the survey lines and tangent lines are extended outward by 5km.
[0043] Airborne gravity measurements were conducted and data was collected according to the established survey lines.
[0044] The raw airborne gravity data collected along the main survey line and the cutting line are processed to obtain airborne free-space gravity anomaly data.
[0045] The specific data processing involves correcting the carrier's motion acceleration, Etherworth correction, eccentricity correction, height correction, normal gravity value correction, zero-point drift correction, and low-pass filtering.
[0046] The low-pass filter parameter is 100s.
[0047] like Figure 3 The image shown is a profile of airborne free-space gravity anomaly obtained by correcting various parameters and applying a 100s low-pass filter to the measured airborne gravity data of a certain survey line in the survey area.
[0048] The second step is to level the free-space gravity anomaly data of the cutting line.
[0049] Search for all intersections between the cutting line and the main survey line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomalies on the cutting line and the main survey line at the intersections, and calculate the discrepancy value of the intersections on the cutting line.
[0050] The measuring points on the cutting line and the main measuring line are fitted using a polynomial fitting method to find the approximate intersection point of the cutting line and the main measuring line. Using the cutting line as a reference, the point on the cutting line closest to the intersection point is searched and taken as the intersection point. The field value of the cutting line at the intersection point remains unchanged, and the field value of the main measuring line at the intersection point is obtained by linear interpolation. The difference between the field value of the cutting line and the field value of the main measuring line is the cross-diff value (FA_cross_diff).
[0051] On the cutting line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points. The result is used as the optimal compensation value for the airborne free space gravity anomaly at each measuring point on the cutting line and is then compensated.
[0052] Median filtering is a nonlinear signal processing technique based on order statistics theory that effectively suppresses noise. It works by replacing the value of a point in a sequence of numbers with the median of all values in its neighboring region. The principle is explained in [link to technical documentation]. Figure 5 As shown.
[0053] The median filtering method based on the moving window is used to filter the discrepancies at intersection points. The calculation formula is as follows:
[0054] yi=Median xi =(xi-v,…xi…, xi+v), i∈Z, v=(n-1) / 2
[0055] Where n is the moving window size, which is an odd number; xi is the discrepancy value at the i-th intersection, and yi is the discrepancy value at the i-th intersection after median filtering based on the moving window. The moving window size n must cover at least the data segment containing three consecutive intersections.
[0056] like Figure 6 As shown, for the intersection of the cutting line and the survey line, the density along the survey line is equivalent to the spacing between the main survey lines, i.e., d≈2km. If the flight speed for airborne gravity measurement is 240km / h and the sampling rate is 10Hz, then the spacing between the measuring points is 6.6m. The density distribution at the intersection points is much sparser than that at the measuring points. Figure 6 Part A in the figure, i.e., the density of measuring points, is approximately 303 times the density of intersection points, see... Figure 6Part B. If a discrepancy value at a certain intersection point is a gross error, to ensure that the compensation value for each measuring point on the cutting line obtained after median filtering based on the discrepancy value is not affected by this gross error, the window size n needs to cover at least the data segment containing the discrepancy value of the three intersection points, i.e., n > 3 × 303. Because the density distribution of intersection points is much sparser than that of measuring points, the filtered curve has a jagged edge. Therefore, further smoothing processing is required after median filtering, and then the median value is used as the optimal correction value to compensate for the data of the cutting line, thus completing the cutting line leveling.
[0057] like Figure 7 As shown, a low-pass filter is used to smooth the median filtering result.
[0058] The third step is to level the free-space gravity anomaly data of the main survey line.
[0059] Search for all intersections between the main survey line and the cutting line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomalies at the intersections on the main survey line and the leveled cutting line, and calculate the intersection discrepancy value (FA_cross_diff) on the main survey line.
[0060] The measuring points on the cutting line and the main measuring line are fitted using a polynomial fitting method to find the approximate intersection point of the main measuring line and the cutting line. Using the main measuring line as the reference, the point on the main measuring line closest to the intersection point is searched and taken as the intersection point. The field value of the main measuring line at the intersection point remains unchanged. The field value of the cutting line at the intersection point is obtained by linear interpolation. The discrepancy value at the intersection point is the difference between the field value of the main measuring line and the field value of the cutting line.
[0061] S32. On the main measurement line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points. The result is used as the optimal compensation value for the airborne free space gravity anomaly at the corresponding measurement points on the main measurement line and compensation is performed.
[0062] The median filtering method based on moving windows filters out the discrepancies at intersection points. The principle and formula have been explained previously and will not be repeated here.
[0063] For the intersection points of the main survey line and the tangent line, the density along the survey line is equivalent to the spacing between the tangent lines, i.e., d≈10km. If the flight speed for airborne gravity measurement is 240km / h and the sampling rate is 10Hz, then the spacing between measurement points is 6.6m. The density distribution at the intersection points is much sparser than that at the measurement points. (See...) Figure 6 In part A, the density of measuring points is approximately 1515 times that of the density of intersection points. Figure 6Section C. If a discrepancy value at a certain intersection point is a gross error, to ensure that the compensation value of each measuring point on the cutting line obtained after median filtering based on the discrepancy value is not affected by this gross error value, the window size n needs to cover at least the data segment containing the intersection point, i.e., n>3×1515. Because the density distribution of intersection points is much sparser than that of measuring points, the filtered curve has a jagged shape. Further smoothing processing is required after median filtering, and then it is used as the optimal correction value to compensate the data of the main measuring line, completing the leveling of the main measuring line.
[0064] The median filtering result was smoothed using a low-pass filter method.
[0065] Figure 7 To level the free-space gravity anomaly map of a certain work area, Figure 8 The image shows an airborne free-space gravity anomaly after processing using the leveling method of this invention. Figure 7 The strip anomalies distributed along the central survey line were significantly eliminated after processing.
[0066] The present invention also includes step S3, verification of leveling results: calculating the vertical first derivative of the leveled free space gravity anomaly data and plotting the results to verify the leveling results.
[0067] like Figure 9 As shown, the vertical first derivative of the leveled data is calculated. There are no obvious stripes along the survey line in the vertical first derivative graph, indicating that the data leveling is qualified.
[0068] When using the cutting line method for leveling, the presence of gross errors in the calculated cross-point discrepancies can severely distort the fitted curve for these discrepancies, affecting the data processing results. Gross errors need to be removed before cutting line leveling, but this process is time-consuming and relies heavily on the data processor's experience. This invention addresses these issues by using measured airborne gravity data from the working area. After completing various corrections, median filtering is applied to the cross-point discrepancies between the cutting line and the main measurement line, and between the main measurement line and the cutting line. The median filtering result is used as the optimal compensation value to compensate for the airborne free-space gravity anomalies of the cutting line and the main measurement line, thus completing the cutting line and main measurement line leveling. This process eliminates the need to remove gross errors from the data, saving data processing time and resolving the reliance on the processor's experience during leveling.
Claims
1. A method for leveling airborne gravity data without removing gross errors, characterized in that, Includes the following steps: S1, Data Measurement and Preparation S11, Lay out the main survey line and the cutting line perpendicular to the direction of the main survey line in the work area; S12, conduct airborne gravity measurements and collect data according to the established survey lines; S13, Process the raw airborne gravity data collected on the main survey line and the cutting line to obtain airborne free space gravity anomaly data; S2, Cutting line airborne free space gravity anomaly data leveling S21, search for all intersections between the cutting line and the main survey line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomaly values between the cutting line and the main survey line at the intersections, and calculate the discrepancy value of the intersections on the cutting line; S22, on the cutting line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points, and the result is used as the optimal compensation value for the airborne free space gravity anomaly at each corresponding measuring point on the cutting line. S3, Leveling of Airborne Free-Space Gravity Anomaly Data on Main Survey Line S31, search for all intersections of the main survey line and the cutting line, obtain the coordinates of the intersections and the corresponding airborne free space gravity anomaly values of the main survey line and the leveled cutting line at the intersections, and calculate the discrepancy value of the intersections on the main survey line. S32. On the main measurement line, the median filtering method based on the moving window is used to filter and smooth the discrepancies at the intersection points. The result is used as the optimal compensation value for the airborne free space gravity anomaly at the corresponding measurement points on the main measurement line and compensation is performed.
2. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, It also includes step S4, leveling result verification: calculate the vertical first derivative of the leveled free-space gravity anomaly data and plot the leveling result.
3. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, In step S21, the position of the intersection point between the cutting line and the main measuring line is determined. Using the cutting line as a reference, the measuring point on the cutting line that is closest to the intersection point is searched and the measuring point is taken as the intersection point. The field value of the cutting line at the intersection point remains unchanged, and the field value of the main measuring line at the intersection point is obtained by linear interpolation.
4. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, In step S31, the position of the intersection point between the main measuring line and the leveled cutting line is determined. Using the main measuring line as a reference, the measuring point on the main measuring line that is closest to the intersection point is searched and the measuring point is taken as the intersection point. The field value of the main measuring line at the intersection point remains unchanged, and the field value of the cutting line at the intersection point is obtained by linear interpolation.
5. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, In steps S21 and S31, the measuring points on the cutting line and the main measuring line are fitted using a polynomial fitting method to determine the approximate intersection point of the main measuring line and the cutting line or the approximate intersection point of the cutting line and the main measuring line.
6. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, The median filtering method based on the moving window is used to filter the discrepancies at intersection points. The calculation formula is as follows: yi=Median xi =(xi-v,…xi…, xi+v), i∈Z, v=(n-1) / 2 Where n is the size of the moving window, taking an odd value; xi is the discrepancy value of the i-th intersection point, and yi is the discrepancy value of the i-th intersection point after median filtering based on the moving window.
7. The airborne gravity data leveling method without removing gross errors according to claim 6, characterized in that, The moving window size n must cover at least the data segment containing three consecutive intersections.
8. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, Processing of raw airborne gravity data includes carrier motion acceleration correction, Etherworth correction, eccentricity correction, altitude correction, normal gravity value correction, zero-point drift correction, and low-pass filtering.
9. The airborne gravity data leveling method without removing gross errors according to claim 1, characterized in that, The distance between cutting lines should be 5 to 10 times the spacing between main test lines, and the test line length should be extended by at least one filter length.