Seismic horizon interpretation skip point processing method

Abstract

This invention provides a method for handling jump points in seismic horizon interpretation, comprising: Step 1, inputting seismic horizon interpretation data for any seismic line, detecting faults and segmenting the data to generate each cluster of seismic horizon interpretation data; Step 2, for each cluster of segmented seismic horizon interpretation data obtained in Step 1, detecting interpretation jump points; Step 3, for any jump point detected in Step 2, establishing a multinomial regression model using the data points after removing the jump point, and recalculating the value at the jump point using this model to replace the original jump point value; Step 4, visualizing the corrected horizon data obtained in Steps 2 and 3 with the original data in the form of a line graph, and comparing and evaluating the satisfaction with the jump point handling effect. This method for handling jump points in seismic horizon interpretation solves the problems of detecting, removing, and reassigning outliers in seismic horizon interpretation data, improves the quality of seismic interpretation data, and thus improves the accuracy of geological mapping.

Description

Technical Field

[0001] This invention relates to the field of oil and gas geophysical exploration technology, and in particular to a method for handling jump points in seismic horizon interpretation. Background Technology

[0002] Seismic stratigraphic interpretation is a common task in oil and gas geophysical exploration. Seismic stratigraphic data is often the fundamental data for drawing stratigraphic maps and thickness maps; therefore, the quality of seismic stratigraphic data is a major factor determining the accuracy of these maps. However, due to the stability of interpretation software systems or human factors during the interpretation process, anomalies may occur in individual data points, resulting in seismic stratigraphic interpretation jump points. Typically, spatial domain filtering can reduce the impact of jump point data on stratigraphic structure and thickness mapping. However, as the error between the jump point value and the actual value increases, the filtering results are increasingly affected by the jump point value, until it excessively influences surrounding data points, causing distortion. In this case, it is necessary to consider reassigning jump point values ​​based on the numerical distribution trend of surrounding data points.

[0003] Interpretive data along any given seismic survey line can obviously be treated as a set of "ordered scattered data", where the values ​​of adjacent data points have significant correlations and specific trends.

[0004] Chinese patent application CN201910280939.0 discloses a method and apparatus for full-layer seismic tracking. The method includes: acquiring 3D seismic data; extracting layer extreme points from the 3D seismic data to construct a sample space; dividing the sample space into multiple subspaces with overlapping portions, clustering the layer extreme points in each subspace to obtain layer fragments corresponding to each layer of the 3D seismic data; establishing topological consistency between layer fragments; and fusing the layer fragments corresponding to each layer of the 3D seismic data based on topological consistency to obtain the full-layer seismic tracking result. This invention eliminates the influence of local connectivity between some different layers in 3D seismic data, avoids the layer cross-linking phenomenon during full-layer tracking, and improves the accuracy of layer fragment fusion by establishing topological consistency, resulting in better full-layer tracking performance.

[0005] Chinese patent application CN201410218588.8 discloses an automatic stratigraphic tracking method based on waveform morphology, belonging to the field of oil and gas geophysical exploration. This method utilizes a global search strategy to extract morphological features of seismic waveforms, compare the similarity of two waveforms, and extract salient points on similar waveforms as stratigraphic points, thus achieving automatic stratigraphic tracking. Using this invention, even in low-energy seismic traces, salient points can still be found through waveform morphological features. When encountering faults, it can skip the faults and continue tracking along the initial features.

[0006] U.S. Patent Application No. US13604964 discloses a method for searching for seismic layers in subsurface seismic images. This method specifically includes specifying two points belonging to the sought horizon, iteratively searching for an optimal solution to the seismic horizon equation, particularly by decomposing the solution into two components, a pseudo-continuous component and a skip component.

[0007] The above-mentioned existing technologies are all quite different from the present invention and have failed to solve the technical problem we want to solve. Therefore, we have invented a new method for handling jump points in seismic horizon interpretation. Summary of the Invention

[0008] The purpose of this invention is to provide a seismic horizon interpretation jump point processing method that can handle outliers in seismic horizon interpretation data.

[0009] The objective of this invention can be achieved through the following technical measures: a seismic horizon interpretation jump point processing method, which includes:

[0010] Step 1: Input seismic horizon interpretation data for any survey line, detect faults and segment the data to generate each cluster of seismic horizon interpretation data;

[0011] Step 2: For each cluster of the segmented seismic horizon interpretation data obtained in Step 1, detect interpretation jump points;

[0012] Step 3: For any jump point detected in Step 2, establish a multinomial regression model using the data points after removing the jump points, and use the model to recalculate the value at the jump point to replace the original jump point value.

[0013] Step 4: Visualize the corrected layer data obtained in Steps 2 and 3 with the original data in the form of a line chart, and compare and evaluate the satisfaction with the jump point processing effect.

[0014] The objective of this invention can also be achieved through the following technical measures:

[0015] In step 1, input seismic horizon interpretation data for any survey line, use the DBSCAN clustering algorithm to detect faults, and segment the data at the fault points. If a fault exists, generate two or more clusters, and the data in each cluster is still sorted in trace order.

[0016] In step 1, for the seismic interpretation layer data (N, T0) input according to the seismic line, where N is the seismic trace number and T0 is the corresponding two-way time series, DBSCAN clustering is performed on T0 using parameters (∈, MinPts). By adjusting the parameters (∈, MinPts), the clustering results can accurately detect faults, where ∈ is the radius of the cluster space and MinPts is the neighborhood density threshold. Then, the data (N, T0) is segmented according to the breakpoints, and the resulting clusters are sorted according to the trace order.

[0017] In step 1, the input seismic horizon interpretation data can be either horizontal or vertical lines.

[0018] In step 2, for each cluster of the segmented seismic horizon interpretation data obtained in step 1, the segmented trend comparison method is used to detect interpretation jump points.

[0019] In step 2, for any cluster of data C obtained in step 1 i For any k ordered points selected sequentially, a quadratic polynomial regression model f is first established. k (x)=a2x 2 +a1x+a0 is used to describe the overall trend of the above k points, and the residual r is calculated. k Where a2 is the coefficient of the quadratic term, a1 is the coefficient of the linear term, and a0 is the constant term; next, the first to k points are successively removed, and a quadratic polynomial regression model f is established for the remaining k-1 points. k-1 (x)=b j2 x 2 +b j1 x+b j0 Where j = 1 to k, and calculate the residual r. kj , where r kj To remove the residual after the j-th point, b j2 b j1 b j0 These are the quadratic coefficients, linear coefficients, and constant term of the quadratic polynomial model established after removing the j-th point; finally, the anomaly detection parameter p = r is constructed. k / r kj When p > a, that is, when r k Greater than r kj When the value is a times greater than p, the currently excluded point is considered a jump point, where a is the threshold of the anomaly detection parameter p.

[0020] In step 3, the correction method for seismic horizon interpretation jump points is to use the regression model f established after removing the point in step 2. k-1 (x)=b j2 x 2 +b j1 x+b j0Calculate the t0 value at this point and replace the original value with the calculated value. Then, merge the processed cluster data according to the seismic trace order to obtain the processed seismic horizon data.

[0021] In step 4, visualize the data obtained in steps 2 and 3 after jump point correction and the original data in the form of a line chart, and compare and evaluate whether the jump point processing effect is satisfactory. If not, reselect the parameters and repeat steps 2 and 3. If satisfied, output the processing result.

[0022] In step 4, the seismic horizon data after jump point detection and correction obtained in step 3 is compared with the original horizon data in the form of a line graph. The effect of the seismic horizon interpretation jump point processing method on the interpretation of the original seismic horizon data is evaluated, and it is determined whether the obtained new horizon data meets the requirements. If the jump point processing effect meets the requirements, the processing result is output; otherwise, return to step 2, adjust the threshold a of the anomaly detection parameter p, and repeat steps 3 and 4 until a satisfactory result is obtained.

[0023] The seismic horizon interpretation jump point processing method in this invention solves the problems of detecting, removing, and reassigning outliers in seismic horizon interpretation data, improving the quality of seismic interpretation data and thus enhancing the accuracy of geological mapping. This method detects faults and segments the interpretation data using DBSCAN clustering. It achieves jump point detection by comparing the established trend model with the original trend model point by point. The method then recalculates the two-way time values ​​at the jump points using a local trend model, thereby avoiding the impact of interpretation jump points and noise on data visualization. It also avoids the excessive influence of outlier data points on the smoothing effect when using spatial filtering algorithms to smooth data. Attached Figure Description

[0024] Figure 1 This is a flowchart of a specific embodiment of the seismic horizon interpretation jump point handling method of the present invention;

[0025] Figure 2 This is a comparison diagram of the original seismic horizon data and the seismic horizon data after jump point processing in a specific embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of local trend analysis of any 5 points on the test line to be processed in a specific embodiment of the present invention. Detailed Implementation

[0027] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, 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 invention pertains.

[0028] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, and / or combinations thereof.

[0029] The seismic horizon interpretation jump point handling method of the present invention includes a method for breaking point detection, horizon data segmentation, jump point identification, and jump point value correction based on horizon data generated in seismic interpretation. This method identifies the location of faults on the seismic survey line based on the DBSCAN clustering algorithm and segments the horizon data into different clusters based on the fault locations. For each segmented cluster, jump point detection and correction are performed: for the i-th cluster data C(i), k points are sequentially selected. First, for the above k points, a quadratic polynomial regression model f is established using least squares regression. k (x)=a2x 2 +a1x+a0, calculate the average residual r of the above k points. k Secondly, the first to k points are successively removed, and a quadratic polynomial regression model f is established for the remaining k-1 points. k-1 (y)=b j2 x 2 +b j1 x+b j0 Let j = 1 to k, and calculate the average residual r of the remaining k-1 points after removing the k-th point. kj Construct anomaly detection parameters p = r k / r kj When p > a, where a is the detection threshold; that is, when r i Greater than r kj When the value is a times greater than 'a', the currently excluded point is considered a jump point and must be corrected; fourth, the regression model f is used. k-1 (x)=b j2 x 2 +b j1 x+b j0 The currently excluded jump points are reassigned values; finally, the data before and after processing are plotted as curves to compare the correction effect, and the threshold 'a' is adjusted until a satisfactory result is achieved. This seismic horizon interpretation jump point processing method detects faults and segments the interpretation data through DBSCAN clustering. By comparing the trend model established through point-by-point elimination with the original trend model, the purpose of jump point detection is achieved. The two-way time value at the jump point is recalculated using the local trend model, thereby avoiding the impact of interpretation jump points and noise on the data visualization effect. It also avoids the excessive influence of abnormal data points on the smoothing effect when using spatial filtering algorithms to smooth the data.

[0030] The following are several specific embodiments of the application of the present invention.

[0031] Example 1

[0032] In a specific embodiment 1 of the present invention, the seismic horizon interpretation jump point processing method of the present invention includes:

[0033] Step 1: Input seismic horizon interpretation data for any seismic line. Use the DBSCAN clustering algorithm to detect faults and segment the data at the fault points. If a fault exists, generate two or more clusters, with each cluster still sorted by trace order. For the seismic horizon interpretation data (N, T0) input by seismic line, where N is the seismic trace number and T0 is the corresponding two-way time series, use parameters (∈, MinPts) for DBSCAN clustering on T0. By adjusting the parameters (∈, MinPts), the clustering results can accurately detect faults, where ∈ is the radius of the cluster space and MinPts is the neighborhood density threshold. Then, segment the data (N, T0) by the fault points, and sort the resulting clusters by trace order.

[0034] The input seismic interpretation horizon can be either a horizontal line or a vertical line.

[0035] Step 2: For each cluster of the segmented seismic horizon interpretation data obtained in Step 1, the segmented trend comparison method is used to detect interpretation jump points;

[0036] For any cluster of data C obtained in step 1 i For any k ordered points selected sequentially, a quadratic polynomial regression model f is first established. k (x)=a2x 2 +a1x+a0 is used to describe the overall trend of the above k points, and the residual r is calculated. k Secondly, the first to k points are successively removed, and a quadratic polynomial regression model f is established for the remaining k-1 points. k-1 (x)=b j2 x 2 +b j1 x+b j0 Where j = 1 to k, and calculate the residual r. kj Finally, construct the anomaly detection parameters p = r k / r kj When p > a, that is, when r k Greater than r kj When the value is a times greater than p, the currently excluded point is considered a jump point, where a is the threshold of the anomaly detection parameter p.

[0037] This seismic horizon jump point detection method is also applicable to the detection of outliers in other ordered data sequences.

[0038] Step 3: For any jump points detected in Step 2, establish a multinomial regression model using the data points after removing the jump points, and recalculate the values ​​at the jump points using this model to replace the original jump point values; the correction method for jump points in seismic horizon interpretation is to use the regression model f established after removing the point in Step 2. k-1 (x)=b j2 x 2 +b j1 x+b j0 Calculate the t0 value at this point and replace the original value with the calculated value. Then, merge the processed cluster data according to the seismic trace order to obtain the processed seismic horizon data.

[0039] Step 4: Visualize the data obtained in Steps 2 and 3 after jump point correction with the original data in the form of a line chart, and compare and evaluate whether the jump point processing effect is satisfactory. If not, reselect the parameters and repeat Steps 2 and 3. If satisfied, output the processing result.

[0040] The seismic horizon data obtained in step 3 after jump point detection and correction is compared with the original horizon data in the form of a line graph. The effect of the seismic horizon interpretation jump point processing method on the interpretation of the original seismic horizon data is evaluated, and it is determined whether the obtained new horizon data meets the requirements. If the jump point processing effect meets the requirements, the processing result is output; otherwise, return to step 2, adjust the threshold a of the anomaly detection parameter p, and repeat steps 3 and 4 until a satisfactory result is obtained.

[0041] Example 2

[0042] In a specific embodiment 2 of the present invention, such as Figure 1 As shown, Figure 1 The flowchart below shows the seismic horizon interpretation jump point processing method of the present invention, which includes the following steps:

[0043] In step 101, input the seismic interpretation horizon data according to the survey line. The process then proceeds to step 102.

[0044] In step 102, Figure 2 The image shows a comparison between the original seismic horizon data and the seismic horizon data after jump point processing. (a) shows the original seismic horizon data, (b) shows the seismic horizon data after jump point processing, and (c) shows a comparison of the effects of the original seismic horizon data and the seismic horizon data after jump point processing.

[0045] For example Figure 2Seismic horizon data along a certain seismic survey line is shown in Figure 'a'. The horizontal axis represents the CDP number, and the vertical axis represents the two-way data. DBSCAN clustering is performed using parameters (∈, MinPts), where ∈ is the radius of the cluster space and MinPts is the neighborhood density threshold. By adjusting parameters ∈ = 5.0 and MinPts = 5, the clustering results can accurately detect fault locations, dividing the horizon data into two clusters from earthquakes A and B. The process then proceeds to step 103.

[0046] In step 103, for any 5 ordered points in each of the two clusters of data obtained in step 2, taking the first 5 points of cluster 1 as an example, a quadratic polynomial regression model is first established:

[0047] Y = 0.2186x 2 -204.54x+49309,

[0048] Figure 3 For the local trend analysis of any 5 points on the test line to be processed, (a) is the quadratic polynomial trend fitting of any 5 points, (b) is the quadratic polynomial trend fitting of the remaining 4 points after removing the 1st point, (c) is the quadratic polynomial trend fitting of the remaining 4 points after removing the 2nd point, (d) is the quadratic polynomial trend fitting of the remaining 4 points after removing the 3rd point, (e) is the quadratic polynomial trend fitting of the remaining 4 points after removing the 4th point, and (f) is the quadratic polynomial trend fitting of the remaining 4 points after removing the 5th point.

[0049] like Figure 3 As shown in figure a, the overall trend of the above 5 points is described, and the residual r is calculated;

[0050] Secondly, points 1 through 5 are removed sequentially, and quadratic multinomial regression models are established for the remaining 4 points after removing each point, such as... Figure 3 b- Figure 3 As shown in f, calculate the residual r. j (j = 1~4);

[0051] Finally, the anomaly detection parameter p = r / r is constructed. j When p > a, the point being removed is considered a jump point; in the example above, a = 3, meaning when r > r j When the value is 3 times the value, the currently excluded point is considered a jump point, such as... Figure 3 In (d), the third point that is removed is the skip point. By using the above method, all skip points in the data of this cluster are detected, and the process proceeds to step 104.

[0052] In step 104, for each jump point detected in step 103, the t0 value at that point is calculated using the regression model established after removing that point in step 103, and the calculated value replaces the original value. Then, the processed cluster data are merged according to the seismic trace order to obtain the processed seismic horizon data. The process then proceeds to step 105.

[0053] In step 105, for the corrected seismic horizon data obtained in step 104, such as Figure 2 b. Compare the data with the original stratigraphic data in a visual manner, such as... Figure 2 c. Evaluate the effectiveness of the seismic horizon interpretation skip point processing method on the interpretation of the original seismic horizon data, and determine whether the obtained new horizon data meets the requirements: if the skip point processing effect meets the requirements, output the processing result; otherwise, return to step 103 and adjust the anomaly detection parameter p = r / r. j The threshold 'a' is adjusted until a satisfactory result is obtained.

[0054] Example 3:

[0055] In a specific embodiment 3 of the present invention, the seismic horizon interpretation jump point processing method of the present invention includes:

[0056] In step 101, input the seismic interpretation horizon data for the survey line to be processed. The process then proceeds to step 102.

[0057] In step 102, for the seismic horizon interpretation data input in step 1, DBSCAN clustering is performed using parameters (∈, MinPts), where ∈ is the radius of the cluster space and MinPts is the neighborhood density threshold. Adjusting parameters ∈ = 5.0 and MinPts = 6 ensures that the clustering results can accurately detect fault locations, dividing the horizon data into two clusters based on the fault results. The process then proceeds to step 103.

[0058] In step 103, for any 7 ordered points in each of the two clusters of data obtained in step 2, taking the 3rd to 9th points of the first cluster as an example, the index and value of these 7 points are...

[0059] 1,1175.59;

[0060] 2,1195.61;

[0061] 3,1190.87;

[0062] 4,1196.46;

[0063] 5,1195.38;

[0064] 6,1185.84;

[0065] 7,1212.93;

[0066] First, establish the quadratic polynomial regression model for the above 7 points:

[0067] Y = -0.0237x 2 +3.6535x+1179.1

[0068] By removing the points to be processed one by one, quadratic polynomial trends are established for the remaining 6 points after removing points 1-7, as follows:

[0069] Y1 = 1.7612x 2 -10.316x+1205.6

[0070] Y2 = 2.4671x 2 -11.453x+1203.5

[0071] Y3 = 3.1443x 2 -16.464x+1212

[0072] Y4 = 3.5436x 2 -18.3x+1212

[0073] Y5 = 3.3893x 2 -17.375x + 1211.2

[0074] Y6 = 2.7079x 2 -12.332x + 1205.5

[0075] Y7 = -1.1621x 2 +5.4699x+1189.2

[0076] Construct anomaly detection parameters p = r / r j The residuals are compared to determine whether the removed points are outliers. In the example above, the anomaly detection parameter threshold a = 2.5, meaning that when r is greater than r... j When the value is 2.5 times the value of the data, the currently discarded point is considered a jump point; in the example above, the 6th point is identified as a jump point. Using the above method, all jump points in the recorded cluster data are detected, and the process proceeds to step 104.

[0077] In step 104, for each jump point detected in step 103, the t0 value at that point is calculated using the regression model established after removing that point in step 103, and the calculated value replaces the original value. Then, the processed cluster data are merged according to the seismic trace order to obtain the processed seismic horizon data. The process then proceeds to step 105.

[0078] In step 105, the corrected seismic horizon data obtained in step 104 is compared with the original horizon data in a visual manner to evaluate the effectiveness of the seismic horizon interpretation jump point processing method on the interpretation jump point processing of the original seismic horizon data, and to determine whether the obtained new horizon data meets the requirements: if the jump point processing effect meets the requirements, the processing result is output; otherwise, return to step 103 and adjust the anomaly detection parameter p = r / r. j The threshold 'a' is adjusted until a satisfactory result is obtained.

[0079] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0080] Except for the technical features described in the specification, all other technologies are known to those skilled in the art.

Claims

1. A method for seismic horizon interpretation jump point processing, characterized in that, The seismic horizon interpretation jump point processing method comprises: Step 1, inputting arbitrary line seismic horizon interpretation data, detecting faults and segmenting data to generate each cluster of the seismic horizon interpretation data; Step 2, detecting interpretation jump points for each cluster of the segmented seismic horizon interpretation data obtained in step 1; Step 3, for any jump point detected in step 2, a polynomial regression model is established by using the data points after removing the jump points, and the value of the jump point is recalculated by using the model to replace the original jump point value; Step 4, the corrected horizon data obtained in steps 2 and 3 and the original data are visualized in the form of a broken line graph, and the satisfaction degree of the jump point processing effect is compared and evaluated; In step 1, for the seismic interpretation horizon data (N, T0) input according to the seismic line, where N is the seismic trace number and T0 is the corresponding two-way time series; parameters are selected for T0 ( DBSCAN clustering is performed using MinPts, by adjusting the parameters ( ϵ is the radius of the cluster space, and MinPts is the neighborhood density threshold. Then, the data (N, T0) is segmented according to the breakpoint, and the resulting clusters are sorted according to the channel order. In step 2, for any one cluster of data C obtained in step 1 i , any k ordered points selected in turn, first, a quadratic polynomial regression model f k (x) = a2x 2 + a1x + a0 is established to express the overall trend of the k points, and a residual r k is calculated, where a2 is a quadratic term coefficient, a1 is a linear term coefficient, and a0 is a constant term; second, the first to kth points are sequentially removed, a quadratic polynomial regression model f k-1 (x) = b j2 x 2 + b j1 x + b j0 is established for the remaining k-1 points, where j = 1~k, and a residual r kj is calculated, where r kj is the residual after removing the jth point, b j2 , b j1 , and b j0 are respectively a quadratic term coefficient, a linear term coefficient, and a constant term of the quadratic polynomial model established after removing the jth point; finally, an anomaly detection parameter p = r k / r kj is constructed, where p > a , i.e., when r k is greater than r kj by a times, the current removed point is considered to be a jump point, where a is a threshold value of the anomaly detection parameter p .

2. The seismic horizon interpretation jump processing method of claim 1, wherein, In step 1, the inputted arbitrary line seismic horizon interpretation data is used to detect faults by using the DBSCAN clustering algorithm, and the data is segmented at the breakpoints, and if the faults exist, two or more clusters are generated, and the data in each cluster is still sorted in the order of the traces.

3. The seismic horizon interpretation jump processing method of claim 1, wherein, In step 1, the inputted seismic horizon interpretation data is a horizontal line or a vertical line.

4. The seismic horizon interpretation jump processing method of claim 1, wherein, In step 2, for each cluster of the segmented seismic horizon interpretation data obtained in step 1, a segmented trend comparison method is used to detect interpretation jump points.

5. The seismic horizon interpretation jump processing method of claim 4, wherein, In step 3, the correction method for the skip point of the seismic horizon interpretation is to use the regression model f established after the point is removed in step 2 k-1 (x) = b j2 x 2 +b j1 x+b j0 The t0 value at the point is calculated and the original value is replaced by the calculated value, and then the processed cluster data is merged in the order of the seismic trace to obtain the processed seismic horizon data.

6. The seismic horizon interpretation jump processing method of claim 1, wherein, In step 4, the jump point corrected data obtained in steps 2 and 3 and the original data are visualized in the form of a broken line graph, and the satisfaction degree of the jump point processing effect is compared and evaluated, if not satisfied, the parameters are reselected, steps 2 and 3 are repeated, if satisfied, the processing result is outputted.

7. The seismic horizon interpretation jump processing method of claim 6, wherein, In step 4, the jump point value detected and corrected seismic horizon data obtained in step 3 and the original horizon data are visualized in the form of a broken line graph, the effect of the seismic horizon interpretation jump point processing method on the original seismic horizon data interpretation jump point processing is evaluated, and it is determined whether the obtained new horizon data meets the requirements: if the jump point processing effect meets the requirements, the processing result is outputted. Otherwise, go to step 2, adjust the abnormality detection parameters p of the threshold a and repeat steps 3, 4 until satisfactory results are obtained.

Images