Automatic extraction and quantitative characterization method for modern sedimentary configuration characteristics of single meandering zone

Abstract

The invention discloses an automatic extraction and quantitative characterization method for modern sedimentary configuration features of a single meandering zone. The method comprises the following steps: S1, constructing a spatial data model of modern sedimentary configuration elements of a meandering river; s2, based on a computational geometry method, extracting and quantitatively representing morphological polygons of modern sedimentary configuration elements of the meandering river one by one, and establishing a spatial relationship among the configuration elements; and S3, quantitative characterization of geometric feature parameters of modern sedimentary configuration elements of the meandering river: determining the arc length, length, width, area and curvature of each feature based on the generated geometric figure of each configuration element and the central axis of the meandering river, and realizing quantitative characterization of the geometric features of the configuration elements. According to the method, the geometrical characteristics and various characteristic parameters of the modern sedimentary configuration elements of the meandering river can be automatically extracted from the riverway of the meandering river, the construction of a knowledge base of the modern sedimentary configuration elements of the meandering river is supported, and the graphical and quantitative characterization of the modern sedimentary configuration characteristics of the meandering river is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of computational geometry and quantitative characterization of geological knowledge, and particularly relates to a method for automatically extracting and quantitatively characterizing the modern sedimentary configuration characteristics of a single meandering belt. Background Art

[0002] Rivers are important channels for water to flow from land to lakes and seas, and are also the main geological agents for transporting sediments from land to seas and lakes. During the river transportation process, sedimentation occurs, forming extensive river deposits. The channel sand bodies formed by river deposits can constitute good places for oil and gas accumulation. If the ancient river sand bodies are close to the oil source, they can become oil and gas reservoirs. Ancient river sand bodies can form lithologic oil reservoirs, stratigraphic-lithologic oil reservoirs, and structural-lithologic oil reservoirs.

[0003] Currently, such oil reservoirs are constantly being discovered around the world, and there are also many large oil and gas fields in China with river facies sand bodies as reservoirs. For example, the Jurassic oilfield in Maling, northern Shaanxi, the Gudao and Gudong oilfields within the scope of the Shengli Oilfield, and the Neogene Guantao Formation in the Bohai Bay Basin. Therefore, people are constantly trying to exploit large oil and gas fields with river facies sand bodies as reservoirs. For example, Chinese Patent Application CN104297787A discloses a three-dimensional lithofacies data processing method and device for river facies low-permeability tight sandstone reservoirs, which relates to the technical field of oil and gas development. The method includes: obtaining the main variogram of lithofacies according to the length, width, thickness, and provenance direction parameters of the sand body; obtaining the natural gamma field according to the natural gamma curve and seismic data; establishing a natural gamma model under the constraint of the main variogram of lithofacies through the sequential Gaussian simulation method according to the natural gamma curve and the natural gamma field; generating a sandstone probability volume according to the corresponding relationship between the natural gamma model and the sandstone probability; establishing a three-dimensional training image, and establishing a lithofacies model through the multiple-point geostatistics method according to the three-dimensional training image, the lithofacies data of the well points, and the sandstone probability volume. Another example is Chinese Patent Application CN 107356958A, which discloses a step-by-step seismic facies prediction method for river facies reservoirs based on geological information constraints, including the following steps: geological data evaluation; establishing an analysis histogram, and comprehensively analyzing the superimposed relationship between the histogram and the channel sand bodies in the work area; abstracting a seismic response model; constituting a set of seismic sensitive attributes; identifying seismic patterns using a probabilistic neural network; performing relevant preprocessing operations on seismic attributes; and seismic facies prediction to obtain a seismic facies map.

[0004] As shown above, river facies reservoirs have always been the focus and difficulty in oil and gas exploration and development. In the middle and late stages of oilfield development, quantitatively analyzing the morphology and scale of river facies sedimentary characteristics is of great significance for oil and gas exploration and development.

[0005] Rivers can be classified into meandering rivers, braided rivers, straight rivers, network rivers, etc. according to the channel geometry. Among them, meandering rivers and braided rivers are the key objects of oil and gas geological research. Due to different geological stresses, the sedimentation processes of these two types of rivers are different, resulting in differences in river sedimentary facies.

[0006] For meandering rivers, point bars are the microfacies units where favorable reservoirs develop in meandering river sedimentation. The scale of single-channel sand bodies and the scale of meandering river point bars are key parameters during the oilfield development stage. An important part of these parameters is the expression of surface geometric features. At present, most solutions use remote sensing image data combined with various types of river data collected, and manually identify the river channels through various vectorization software. By manually calibrating the inflection points, confluence points and other features of the river, sedimentary features are extracted, and then the corresponding geometric parameters are calculated. However, this method has a large workload and the recognition results are restricted by the subjective understanding of the staff. At present, a large number of remote sensing map data sources provide rich materials for the study of modern sedimentary configuration features. However, how to efficiently and accurately extract feature parameters from them to form a geological knowledge base in the big data environment is still a challenging problem. Summary of the Invention

[0007] Object of the Invention: In view of this, the present invention provides a method for automatically extracting and quantitatively characterizing the modern sedimentary configuration features of a single meander belt, so as to provide an intelligent and efficient technical means for the extraction of modern sedimentary configuration features of meandering rivers and the construction of their knowledge bases in the current big data environment.

[0008] Technical Solution: A method for automatically extracting and quantitatively characterizing the modern sedimentary configuration features of a single meander belt includes the following steps:

[0009] S1. Construct a spatial data model of modern sedimentary configuration elements of meandering rivers, where:

[0010] The modern sedimentary configuration elements of meandering rivers include meander belts, curved river reaches, and point bars;

[0011] S2. Based on computational geometry methods, extract and quantitatively characterize the morphological polygons of modern sedimentary configuration elements of meandering rivers one by one, and establish the spatial relationships between various configuration elements;

[0012] S3. Quantitative characterization of geometric feature parameters of modern sedimentary configuration elements of meandering rivers:

[0013] Based on the geometric shapes of the generated configuration elements and their meandering river centerlines, determine the arc length, length, width, area, and curvature of each feature, and realize the quantitative characterization of the geometric features of the configuration elements.

[0014] Furthermore, the specific steps of step S1 are as follows:

[0015] S11. Describe and define the geometric shapes of the configuration elements of the meander belt, meandering reach, and point bar;

[0016] S12. Describe and define the characteristic parameters of the configuration elements of the meander belt, meandering reach, and point bar.

[0017] Furthermore, in step S11, the geometric shapes of the meander belt, meandering reach, and point bar are all polygon objects, where:

[0018] There is an inclusion relationship between the meander belt and the meandering reach;

[0019] There is an inclusion relationship between the meander belt and the point bar;

[0020] There is an adjacency relationship between the meandering reach and the point bar;

[0021] There is an adjacency relationship between adjacent meandering reaches.

[0022] Furthermore, in step S12, the characteristic parameters of the configuration elements of the meander belt, meandering reach, and point bar all include arc length, length, width, area, and curvature.

[0023] Further, the computational geometry method described in step S2 is the Voronoi diagram skeleton line.

[0024] Further, the specific steps of step S2 are as follows:

[0025] S21. Generate the meander belt polygon;

[0026] S22. Subdivide the meander belt polygon and extract the meandering river centerline;

[0027] S23. Determine the candidate inflection point set of the meandering river centerline based on the vector rotation direction, and filter out the unreasonable inflection points in the candidate inflection point set of the meandering river centerline through the meandering reach scale, river morphology, and the periodic law of the meander river bends;

[0028] S24. Connect the identified inflection points of the meandering river centerline in sequence to form the entire meander belt centerline; draw transverse lines perpendicular to the meander belt centerline through each inflection point to divide the meandering river channel into sections of meandering reaches;

[0029] S25. Point bar polygon generation:

[0030] For each meandering reach, generate the corresponding point bar polygon for the region enclosed by the line segment connecting its two inflection points and its corresponding convex bank boundary line, and record the adjacency relationship between the meandering reach and the point bar.

[0031] Furthermore, the specific steps of step S21 are as follows:

[0032] S211. Construct a Delaunay triangulation for the meandering river channel area;

[0033] S212. Perform a "peeling" operation on the Delaunay triangulation, that is, gradually delete the triangles located on the periphery with side lengths greater than the first threshold, erode layer by layer inward until the side lengths of the triangles on the outer boundary are all less than the first threshold. At this time, the polygon formed by connecting the outer triangle sides in sequence constitutes the distribution range of the meander belt, where:

[0034] Define the filtering factor of the Delaunay triangulation as k, and the average side length of the triangles in the Delaunay triangulation as l avg , where:

[0035] The first threshold is k * l avg , that is, when there is a side with a length greater than k * l among the three sides of a triangle avg , then the triangle is removed from the triangulation;

[0036] S213. Merge the remaining triangles in the Delaunay triangulation to obtain the meander belt polygon.

[0037] Furthermore, step S22 includes the following steps:

[0038] S221. Construct a Voronoi diagram skeleton line with the inflection points of the outer contour line of the formed meander belt polygon as the point set;

[0039] S222. Convert the Voronoi diagram skeleton line into the expression form of an undirected graph, where:

[0040] Define an undirected graph G(V, E), where: V is the set of intersection points of Voronoi polygons in the Voronoi diagram {v i | i = 1,..., M}, and E is the set of sides of Voronoi polygons in the Voronoi diagram {e j | j = 1,..., N}, and neither V nor E contains duplicate objects;

[0041] S223. Define the set of sides constituting the meandering river axis as C = {e k ,..., e l}, such that is minimized. Based on this, use the shortest path algorithm to search the undirected graph G(V, E) to find the target set of edges C, and this set C can be used to represent the meandering river axis.

[0042] Even further, the specific steps of step S221 are as follows:

[0043] S2211. Divide the irregular triangulated network that meets the Delaunay criterion from the vertex points of the outer contour lines of all meander belt polygons;

[0044] S2212, the edges of Thiessen polygons can be formed by the perpendicular bisectors of the triangle edges. The intersection of the bisectors determines the position of the Voronoi edge inflection points. The geometric network formed by the edges and intersections of the Voronoi polygons inside the meander belt polygon is the skeleton line of the Voronoi diagram.

[0045] Furthermore, the specific steps of step S23 are as follows:

[0046] S231, determine the candidate inflection point set of the central axis based on the vector rotation direction;

[0047] S232, filter the turning points in the candidate turning point set of the central axis based on the scale of the curved river section:

[0048] The fluctuation of the central axis shape will lead to the extraction of multiple inflection points within a short distance. At this time, the number of central axis vertices between the extracted inflection points is generally small. Therefore, the curved river section scale threshold s is set. After obtaining the candidate inflection point set, the number of central axis vertices between every two inflection points is calculated, and the inflection points with the number of vertices less than the curved river section scale threshold s are directly deleted;

[0049] S233, filter the turning points in the candidate turning point set of the central axis based on the river morphology:

[0050] When the river section between adjacent turning points in the central axis candidate turning point set appears to be approximately straight, delete the corresponding turning points;

[0051] S234, filtering the turning points in the candidate turning point set of the central axis based on the periodic law of the meandering river bend:

[0052] In the same curved section of a meandering river, there are often only two inflection points. The polygonal set formed by the inflection point connection line and the central axis is defined as T = {t m |m=1,...,O}, at this time, the combination mode of adjacent river sections in T can be divided into two types: B-type mode and S-type mode, where:

[0053] The B-type mode means that the adjacent polygons are located on one side of the central axis, and the S-type mode means that the adjacent polygons are located on both sides of the central axis. Only the inflection points of the S-type mode are retained, and the inflection points of the B-type mode are eliminated, where:

[0054] Assume that the vertex set on the meandering river central axis in S223 is expressed as L = {p 1 ,p 2 , ..., p m}, where P j ​(j ∈ 1, 2, ..., m) is the j-th vertex on the axis of the meandering river, P k (k ∈ 1, 2, ..., m) is the k-th vertex on the axis of the meandering river, then the method for judging the combined pattern of adjacent curved river reaches is as follows:

[0055] A, B, and C are any three adjacent inflection points on the axis of the meandering river. From inflection point A to the vertex P between inflection points A and C j make a vector Similarly, from inflection point B to the vertex P between inflection points B and C k make a vector Let When the vectors are in opposite directions, inflection point B is an S-mode inflection point. When the vectors are in the same direction, inflection point B is a B-mode inflection point. At this time, delete inflection point B, and then continue to judge the next adjacent curve until all inflection points are traversed.

[0056] Furthermore, the specific steps of step S231 are as follows:

[0057] S2311: Extract all the vertices on the axis of the meandering river. Take the direction corresponding to the default storage order of the line segments in the vector of the axis of the meandering river as the direction of the axis of the meandering river. Take the starting point of the axis of the meandering river as the tail of the first vector, and the next point of the starting point as the head of the first vector. Traverse the vertices on the axis of the meandering river in turn to construct vectors until all vertices are traversed and the loop ends. At this time, the vector group corresponding to the axis of the meandering river is obtained;

[0058] S2312: Calculate the change in the direction of the front and rear vectors in this vector group. When the change direction of the angle between the front and rear vectors changes, this point is considered to be an inflection point in the geometric sense.

[0059] Furthermore, the specific steps of step S3 are as follows:

[0060] S31: Let the polygon representation of the meander belt or a curved river reach be P = {v 1 , v 2 , …, v n}, where v i (i ∈ 1, 2, …, n) is the i-th vertex on the boundary line, and its coordinates are (x i , y i ); Since the polygon boundary line is closed, the coordinates of v 1 and v n coincide; its axis of the meandering river is defined as L = {p 1 , p 2 , …, p m}, where p j(j ∈ 1, 2, …, m) is the j-th node on the meandering river axis, and its coordinates are (x j , y j ). Then the calculation methods of the basic geometric parameters are as follows:

[0061] ①. Length of the meandering river axis:

[0062] ②. Straight-line length:

[0063] ③. Area:

[0064] ④. Average width:

[0065] ⑤. Curvature:

[0066] ⑥. Perimeter:

[0067] S32. For a point bar, its length and width are different from those of the bending river reach:

[0068] Let the two inflection points forming the point bar be point v k , v l , and their coordinates are (x k , y k ) and (x l , y l ). Then the calculation formula for the point bar length is:

[0069] ⑦. Point bar length:

[0070] Set the vertex set on the convex bank arc enclosing the point bar as P = {p 1 , p 2 , …, p n}. Then a perpendicular line is drawn from point p i to the line segment v k v l , and the foot of the perpendicular is O. The coordinates of point p i and O are (x i , y i ) and (x o , y o ). The distance between p i and O is Then the point bar width can be determined as:

[0071] ⑧. Point bar width: w = MAX(||p i O||), where MAX represents taking the maximum value, and i ∈ {1, 2, …, n}.

[0072] Beneficial effects: The method for automatically extracting and quantitatively characterizing the modern sedimentary configuration features of a single meander belt provided by the present invention has the following beneficial effects:

[0073] Using this method, the geometric characteristics and various characteristic parameters of the modern sedimentary configuration elements of meandering rivers can be automatically extracted from meandering river channels, supporting the construction of a knowledge base for modern sedimentary configuration elements of meandering rivers, and realizing the graphical and quantitative characterization of modern sedimentary configuration features of meandering rivers. Description of the drawings

[0074] Figure 1 It is a flow chart of the method for automatically extracting and quantitatively characterizing the modern sedimentary configuration features of a single meander belt disclosed by the present invention.

[0075] Figure 2 It is a schematic diagram of the spatial data model of the modern sedimentary configuration elements of meandering rivers constructed in step S1.

[0076] Figure 3 It is an entity relationship diagram of the spatial data model of the modern sedimentary configuration elements of meandering rivers constructed in step S1.

[0077] Figures 4a - 4c It is the method for identifying meander belts in step S21.

[0078] Figure 5 It is a schematic diagram of the method for determining inflection points in step S23.

[0079] Figure 6 It is a schematic diagram of the meandering river centerline in step S23.

[0080] Figure 7 It is a schematic diagram of filtering inflection points based on river morphology in step S23;

[0081] Figure 8a It is a schematic diagram of inflection points of the B-type pattern.

[0082] Figure 8b It is a schematic diagram of inflection points of the S-type model.

[0083] Figure 9a It is a schematic diagram of the method for judging inflection points of the B-type pattern.

[0084] Figure 9b It is a schematic diagram of the method for judging inflection points of the S-type pattern.

[0085] Figure 10 It is the method for dividing the meandering reach and determining the point bar range in steps S24 and S25.

[0086] Figure 11 It is a schematic diagram of the method for calculating the point bar length and point bar width in step S31.

[0087] Figure 12 The test data range used for verifying the automatic extraction and quantitative characterization method of the modern deposition configuration characteristics of a single meander belt of the present invention.

[0088] Figures 13a - 13e The test results of the automatic extraction and quantitative characterization method of the modern deposition configuration characteristics of a single meander belt of the present invention in the Baihe meander belt test set.

[0089] Figure 14 Schematic diagram of the statistical results of some characteristic parameters of the geometric figure of the curved reach in the test data.

[0090] Figure 15a Schematic diagram of the correlation analysis of the point bar length and the point bar width when the curvature is less than 1.7.

[0091] Figure 15b Schematic diagram of the correlation analysis of the point bar length and the point bar width when the curvature is greater than or equal to 1.7. Detailed implementation manners

[0092] The following specifically describes the specific embodiments of the present invention with reference to the accompanying drawings. Among them, the drawings in the specification form a part of this application and are used together with the embodiments of the present invention to explain the principle of the present invention, but are not used to limit the scope of the present invention.

[0093] Embodiment 1

[0094] Figure 1 It is the flow chart of the automatic extraction and quantitative characterization method of the modern deposition configuration characteristics of a single meander belt disclosed by the present invention. As Figure 1 shown, the automatic extraction and quantitative characterization method of the modern deposition configuration characteristics of a single meander belt includes the following steps:

[0095] S1. Construct a spatial data model of the modern deposition configuration elements of a meandering river:

[0096] As Figure 2 shown, the spatial data model of the modern deposition configuration elements of a meandering river describes three configuration elements: meander belt, curved reach, and point bar. The geometric figure of each configuration element is a polygon object, where:

[0097] cc is the meander belt polygon;

[0098] c1 - c6 are the curved reach polygons;

[0099] b1 - b6 are the point bar polygons.

[0100] There are two spatial relationships between these polygon objects: ① The inclusion relationship between the meander belt and the curved reach, and the meander belt and the point bar; ② The adjacency relationship between the curved reach and the point bar, and adjacent curved reaches.

[0101] Figure 3 It is the entity relationship diagram of the spatial data model of the modern sedimentary configuration elements of the meandering river constructed in step S1. Figure 3 The definitions of configuration elements such as meander belt, curved river section and point bar and their attributes and relationships are given in the form of entity-relationship diagram, among which the configuration characteristic parameters of each entity mainly include arc length, length, width, area, curvature, etc.

[0102] S2. Based on computational geometry methods such as the Voronoi diagram skeleton line, the morphological polygons of various configuration elements of modern sedimentary structures of meandering rivers are extracted and quantitatively characterized, and the spatial relationship between the elements is established.

[0103] S21, generate meander belt polygon:

[0104] If Figure 4a As shown in , a Delaunay triangulation network is constructed for the meandering river channel. The generated triangulation network will cover the entire meander belt area, but the coverage of the triangulation network is often larger than the actual coverage of the meander belt. The side length of the triangle outside the meander belt is much longer than the triangle inside the meander belt. Therefore, only when the side length of the triangle is less than the visual proximity distance, its connection can serve as the boundary of the meander belt.

[0105] If Figure 4b As shown, the Delaunay triangulation is subjected to a "skinning" operation, that is, gradually deleting those triangles whose sides are greater than a certain distance from the periphery, and eroding inward layer by layer until the sides of the triangles on the periphery are less than a certain threshold. At this time, the polygons obtained by sequentially connecting the sides of the periphery triangles constitute the distribution range of the meander belt. Define the triangulation filter factor k, and the average side length of the triangles in the triangulation is l avg . When one of the three sides of the triangle has a length greater than k*l avg When the edge of the triangle is removed from the triangulated network, k is set to 3 in this embodiment.

[0106] Finally, if Figure 4c As shown in , merging the remaining triangles in the triangulated network can obtain meander belt polygons.

[0107] S22, dividing the meander belt polygon and extracting the meander river central axis:

[0108] The meandering river axis is the line connecting the midpoints of the outer and inner bank edges of all cross-sections of the river channel, which is an abstract conceptual line; each point on the line can represent the planar geometric center of a certain cross-section of the river channel. When conducting linear and morphological studies on the river channel, the meandering river axis can be approximately regarded as the river channel itself. The use of this concept makes the study of the river channel streamline and migration trajectory more convenient and clear, without considering the differences in the distances between the outer and inner banks. By extracting the meandering river axis, the morphological characteristics of the meandering river can be further explored. The present invention realizes the extraction of the meandering river axis by means of the Voronoi diagram skeleton line. The main steps are as follows:

[0109] S221. First, construct the Voronoi diagram skeleton line with the inflection points forming the outer contour line of the meandering river as the point set. Its construction process is as follows:

[0110] First, divide an irregular triangular mesh that conforms to the Delaunay criterion among all the points;

[0111] Then, the sides of the Thiessen polygons can be formed by the perpendicular bisectors of the sides of the triangles. The intersection points of the bisectors determine the positions of the inflection points of the Voronoi polygons. The geometric network formed by the sides and intersection points of the Voronoi polygons inside the river channel polygon is the Voronoi diagram skeleton line;

[0112] S222. To facilitate the search for the meandering river axis, it is necessary to convert the Voronoi diagram skeleton line into an undirected graph representation form. Specifically, define an undirected graph G(V, E), where V is the set of intersection points of Voronoi polygons in the Voronoi diagram {v i | i = 1,..., M}, and E is the set of sides of Voronoi polygons in the Voronoi diagram {e j | j = 1,..., N}, and neither V nor E contains duplicate objects. Based on the undirected graph G(V, E), the problem of extracting the meandering river axis can be transformed into a path search problem in the undirected graph;

[0113] S223. Define the set of sides constituting the meandering river axis as C = {e k ,..., e l}, such that is minimized. Based on this, use the shortest path algorithm to search the graph G(V, E) to find the set C of target sides, and this set C can be used to represent the meandering river axis. Among them, the shortest path search algorithm is a classic algorithm in graph theory and will not be elaborated here.

[0114] S23. Extraction of inflection points. An inflection point is the transitional position between two adjacent curved river reaches. Morphologically, it is the intersection point of the axial line and the center line of the meander belt. The appearance of an inflection point indicates a change in the migration direction of the river channel. Based on the line segments of the center line of the meandering river, vectors are constructed. All candidate inflection points are extracted from the center axis of the meandering river based on the change in vector direction. Then, through multiple steps of filtering based on the morphological characteristics of the meandering river and the global periodic law, etc., the optimal set of inflection points is selected from the candidate inflection points. The main steps are as follows:

[0115] S231. Determine the candidate inflection point set of the center axis of the meandering river based on the vector rotation direction. Extract all the vertices of the center axis of the meandering river. Take the direction corresponding to the default storage order of the line segments in the vector of the center axis of the meandering river as the direction of the center axis of the meandering river. Take the starting point of the center axis of the meandering river as the tail of the first vector, and the next point of the starting point as the head of the first vector. Traverse the vertices on the center axis of the meandering river in this way to construct vectors until all vertices are traversed and the loop ends. At this time, the vector group corresponding to the center axis of the meandering river can be obtained. Calculate the change in the direction of the front and rear vectors in this vector group. When the change direction of the angle between the front and rear vectors changes, this point can be considered an inflection point in the geometric sense. As Figure 5 shown, before and after vertex A, the azimuth angles of the vectors show a monotonically increasing and monotonically decreasing trend respectively. A similar pattern exists at point B. Therefore, vertices A and B are candidate inflection points. However, the extracted center axis of the meandering river often has small undulations (as Figure 6 shown). Therefore, simply using the change in vector direction to determine inflection points is sensitive to the undulations of the center axis of the meandering river, resulting in the number of candidate inflection points obtained at this time being more than the actual morphological inflection points of the meandering river. They cannot be directly used as the inflection points of the river. Therefore, the following steps of filtering are required:

[0116] S232. Filter inflection points based on the scale of the curved river reach. The undulations in the morphology of the center axis of the meandering river will result in the extraction of multiple inflection points within a short distance. Generally, there are fewer vertices of the center axis of the meandering river between the extracted inflection points at this time. Therefore, a threshold S for the scale of the curved river reach is set. After obtaining the candidate inflection point set, calculate the number of vertices of the center axis of the meandering river between every two inflection points, and directly delete the inflection points with the number of vertices less than s. In this embodiment, the value of S is set to 8.

[0117] S233. Filter inflection points based on the river morphology. In a meandering river, there is no completely straight river reach. When the river reach between adjacent inflection points in the candidate set appears to be approximately straight, the corresponding inflection points should be deleted. Specifically, as Figure 7 shown, connect any two adjacent inflection points A and B to form a straight line AB. Obtain all the vertices of the center axis of the meandering river between the two inflection points A and B, draw a perpendicular line to the straight line AB, and take out the perpendicular line OC with the longest length. Calculate The larger the value of K, the closer the river is to being straight. Discard the inflection points with a value greater than the threshold. In this embodiment, the threshold is set to 8.

[0118] S234. Filter inflection points based on the periodic law of meandering river bends. In the same meandering river reach, there are usually only two inflection points. Define the polygon set formed by the connection line of the inflection points and the central axis of the meandering river as T = {t m |m = 1,..., O}. At this time, the combination patterns of adjacent river reaches in T can be divided into two types: B-type pattern and S-type pattern (as shown in Figure 8a , Figure 8b , Figure 9a , Figure 9b ). The B-type pattern means that adjacent polygons are located on one side of the central axis of the meandering river, and the S-type pattern means that adjacent polygons are respectively on both sides of the central axis of the meandering river. In this embodiment, only the S-type pattern is retained, and the inflection points of the B-type are removed. Let the vertex set on the central axis of the meandering river obtained in S223 be represented as L = {p 1 , p 2 ,..., p m}, where P j (j ∈ 1, 2,..., m) is the j-th vertex on the central axis of the meandering river, and P k (k ∈ 1, 2,..., m) is the k-th vertex on the central axis of the meandering river. Then the method for judging the combination pattern of adjacent meandering river reaches is as shown in Figure 9a and Figure 9b . A, B, and C are any three adjacent inflection points. Make a vector j from inflection point A to the vertex P between inflection points A and C. Similarly, make a vector k from inflection point B to the vertex P between inflection points B and C. Let When the directions of the vectors are opposite, inflection point B is an S-mode inflection point. When the directions of the vectors are the same, inflection point B is a B-mode inflection point. At this time, inflection point B should be deleted, and then continue to judge the next adjacent meander until all inflection points meet the requirements.

[0119] S24. Generation of the central axis of the meandering river and division of meandering river reaches. As shown in Figure 10 , connect the inflection points of the central axis of the meandering river identified in the previous step in sequence to form the central axis of the entire meander belt; draw transverse lines perpendicular to the central axis of the meandering river through each inflection point to divide the meandering river channel into sections of meandering river reaches.

[0120] S25. Generation of point bar polygons. Point bars are formed during the lateral migration of meandering rivers and the continuous deposition on the convex bank. As shown in Figure 10As shown in the figure, for a meandering river reach, the area enclosed by the line segment connecting the two inflection points and the corresponding convex bank boundary line can be approximated as the corresponding point bar range. Based on this, the corresponding point bar polygon is generated, and the adjacency relationship between the meandering river reach and the point bar is recorded.

[0121] S3. Quantitative characterization of geometric feature parameters of modern sedimentary architecture elements in meandering rivers. Based on the geometric shapes of the architecture elements and the meandering river axis, basic feature parameters such as arc length, length, width, area, and curvature of various objects are determined to achieve the quantitative characterization of the geometric features of the architecture elements.

[0122] S31. By the above method, the meander belt, meandering river channel polygon, and meandering river axis are determined. They are all simple polygons, and the meandering river axis is a polyline. Based on this graphic information, their planar geometric feature parameters can be further determined. Let the polygon representing the meander belt or a meandering river reach be P = {v 1 , v 2 , …, v n}, where v i (i ∈ 1, 2, …, n) is the i-th vertex on the boundary line, and its coordinates are (x i , y i ). Since the polygon boundary line is closed, the coordinates of v 1 and v n coincide; its meandering river axis is defined as L = {p 1 , p 2 , …, p m}, where p j (j ∈ 1, 2, …, m) is the j-th node on the meandering river axis, and its coordinates are (x j , y j ). Then the calculation methods of the basic geometric parameters are as follows:

[0123] ①. Length of the meandering river axis:

[0124] ②. Straight-line length:

[0125] ③. Area:

[0126] ④. Average width:

[0127] ⑤. Curvature:

[0128] ⑥. Perimeter:

[0129] S32. As Figure 11As shown in the figure, for point bars, their length and width are different from those of meandering river reaches and need special consideration. Let the two inflection points that make up the point bar be point v k , v l , whose coordinates are (x k , y k ) and (x l , y l ), then the calculation formula for the length of the point bar is:

[0130] ⑦. Length of point bar:

[0131] Set the vertex set on the convex bank arc enclosing the point bar as P = {p 1 , p 2 , …, p n}, then draw a perpendicular line from point p i to the line segment v k v l , and the foot of the perpendicular is O. The coordinates of point p i and O are (x i , y i ) and (x o , y o ) respectively. The distance between p i and O is Then the width of the point bar can be determined as:

[0132] ⑧. Width of point bar: w = MAX(||p i O||), where MAX represents taking the maximum value, and i ∈ {1, 2, …, n}.

[0133] As described above, this is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

[0134] Example 2

[0135] To verify the effectiveness of the method of this invention patent application, based on the basic functions provided by the open-source GIS platform QGIS, a toolkit for automatic extraction and quantitative characterization of the modern sedimentary architecture characteristics of meandering rivers was further developed.

[0136] Select a part of the water area of the Baihe River, and use QGIS for vectorization as test data to verify the applicability of the relevant method in actual application scenarios. The range of the test data is as Figure 12 shown.

[0137] The Baihe River is a meandering river formed by natural development, with well-developed vegetation. The geometric shape of its river channel is winding, making it an ideal object for quantitatively studying the single meander belt and its planar geometric characteristics. The Baihe River is located in the Tethys-Himalayan tectonic domain, so the neotectonic movement is very strong. On both sides of the upper reaches are hilly landforms, the terrain in the middle reaches becomes gentler, and the lower reaches enter the Zoige Basin formed by the uplift of the Qinghai-Tibet Plateau. Generally, a narrow and straight river valley is formed in the upper reaches, and a broad linear river valley is formed in the middle and lower reaches. The area of the Baihe alluvial plain increases from small to large. The Baihe River has arc-shaped, strip-shaped, U-shaped, and slightly curved point bars, and the shape of the point bars tends to be regular along the way; there are Ω-shaped, U-shaped, and crescent-shaped oxbow lakes, among which the number of crescent-shaped oxbow lakes is the largest, the U-shaped ones are the second, and the number of Ω-shaped ones is the smallest. The number of oxbow lakes shows a decreasing trend along the way.

[0138] Figures 13a - 13e These are the test results of the automatic extraction and quantitative characterization method for the modern sedimentary configuration characteristics of the single meander belt of the present invention in the Baihe River meander belt test set. As Figures 13a - 13e shown, the extraction results are basically consistent with the remote sensing images and general geological understandings.

[0139] The statistical results of some characteristic parameters of the geometric figures of the test data point bars are shown in the following table:

[0140]

[0141] Figure 14 This is a schematic diagram of the statistical results of some characteristic parameters of the geometric figures of the meandering reaches of the test data. Figure 15a This is a schematic diagram of the correlation analysis of the point bar length and point bar width when the curvature is less than 1.7. Figure 15b This is a schematic diagram of the correlation analysis of the point bar length and point bar width when the curvature is greater than or equal to 1.7. Figure 14 、 Figure 15a 、 Figure 15b respectively show the statistical results of some characteristic parameters of the geometric figures obtained from the segmentation of the meandering reaches and the extraction of point bars in the test data. The relevant results conform to the basic characteristics of the study area and can meet the correctness requirements in practical applications.

[0142] The above has made a detailed description of the implementation manner of the present invention. However, the present invention is not limited to the above implementation manner, and various changes can be made without departing from the purpose of the present invention within the knowledge scope of those of ordinary skill in the art.

Claims

1. Automatic extraction and quantitative characterization method of modern sedimentary configuration features of a single meander belt, characterized by: The following steps are involved: S1. Construct a spatial data model of modern sedimentary configuration elements of meandering rivers, including: The modern sedimentary configuration elements of the meandering river include meander belts, curved river sections, and point bars; S2. Based on computational geometry methods, the morphological polygons of modern sedimentary configuration elements of meandering rivers are extracted and quantitatively characterized one by one, and the spatial relationship between each configuration element is established; S3. Quantitative characterization of geometric characteristic parameters of modern sedimentary architecture elements of meandering rivers: Based on the generated geometric figures of each configuration element and the central axis of the meandering river, the arc length, length, width, area, and curvature of each feature are determined to achieve quantitative characterization of the geometric characteristics of the configuration elements.

2. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 1, characterized in that: The specific steps of step S1 are as follows: S11. Describe and define the geometry of the configuration elements of meanders, curved reaches, and point bars; S12. Describe and define the characteristic parameters of the configuration elements of meander belts, curved river sections, and point bars.

3. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 2, characterized in that: In step S11, the geometric figures of meander belt, curved river section and point bar are all polygonal objects, where: There is an inclusion relationship between the meander belt and the curved river section; There is an inclusion relationship between meander belt and point bar; There is an adjacent relationship between the curved river section and the point dam; Adjacent curved river sections are in an adjacent relationship.

4. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 2, characterized in that: In step S12, the characteristic parameters of the configuration elements of the meander belt, the curved river section, and the point bar all include arc length, length, width, area, and curvature.

5. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 1, characterized in that: The computational geometry method in step S2 is the Voronoi diagram skeleton line.

6. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 1, characterized in that: The specific steps of step S2 are as follows: S21, generating meander belt polygons; S22, dividing the meander belt polygon and extracting the meander river central axis; S23, determining a set of candidate turning points of the meandering river central axis based on the vector rotation direction, and filtering out unreasonable turning points in the set of candidate turning points of the meandering river central axis according to the scale of the meandering river section, the river morphology, and the periodicity of the meandering river bends; S24, connecting the identified turning points of the meandering river central axis in sequence to form the central axis of the entire meandering belt; making a cross-section perpendicular to the central axis of the meandering belt through each turning point to divide the meandering river channel into sections of curved river segments; S25. Point dam polygon generation: For each curved river section, the corresponding point-dam polygon is generated by the line segment connecting its two inflection points and the area enclosed by its corresponding convex bank boundary line, and the adjacency relationship between the curved river section and the point-dam is recorded.

7. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 6, characterized in that: The specific steps of step S21 are as follows: S211. Construct a Delaunay triangulation network for meandering river areas; S212, performing a "peeling" operation on the Delaunay triangulation, that is, gradually deleting triangles located on the periphery whose side lengths are greater than the first threshold, and eroding inward layer by layer until the side lengths of the triangles on the periphery boundary are all less than the first threshold. At this time, the polygons obtained by sequentially connecting the periphery triangle sides constitute the distribution range of the meander belt, where: Define the filter factor of the Delaunay triangulation as k, and the average side length of the triangles in the Delaunay triangulation as l avg ,in: The first threshold is k*l avg , that is, when there is a length greater than k*l among the three sides of the triangle avg When the edge of , the triangle is removed from the triangulated network; S213. Merge the remaining triangles in the Delaunay triangulation to obtain meander belt polygons.

8. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 6, characterized in that: Step S22 includes the following steps: S221, constructing a Voronoi diagram skeleton line using the inflection points of the outer contour line of the meander belt polygon as a point set; S222, converting the Voronoi diagram skeleton line into an expression form of an undirected graph, wherein: Define an undirected graph G(V, E), where: V is the set of intersections of the Voronoi polygons in the Voronoi diagram {v i |i=1,...,M}, E is the set of Voronoi polygon edges in the Voronoi diagram {e j |j=1,...,N}, neither V nor E contains duplicate objects; S223. Define the set of edges that form the central axis of the meandering river as C = {e k , ..., e l }, so that Based on this, the shortest path algorithm is used to search the undirected graph G(V, E) to find the set C of target edges, which can be used to represent the central axis of the meandering river.

9. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 8, characterized in that: The specific steps of step S221 are as follows: S2211. Divide the points of the outer contours of all meander polygons into lines that conform to Delauna y Criteria for triangulated irregular networks; S2212. The edges of Thiessen polygons can be formed by the perpendicular bisectors of the sides of the triangle. The intersection of the bisectors determines the position of the Voronoi edge inflection points. The geometric network formed by the edges and intersections of the Voronoi polygons inside the meander belt polygon is the skeleton line of the Voronoi diagram.

10. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 6, characterized in that: The specific steps of step S23 are as follows: S231, determining a set of candidate inflection points of the central axis based on the vector rotation direction; S232, filtering the turning points in the candidate turning point set of the central axis based on the scale of the curved river section: The fluctuation of the central axis shape will lead to the extraction of multiple inflection points within a short distance. At this time, the number of central axis vertices between the extracted inflection points is generally small. Therefore, the curved river section scale threshold s is set. After obtaining the candidate inflection point set, the number of central axis vertices between every two inflection points is calculated, and the inflection points with the number of vertices less than the curved river section scale threshold s are directly deleted. S233, filtering the inflection points in the candidate inflection point set of the central axis based on the river morphology: When the river section between adjacent turning points in the central axis candidate turning point set appears to be approximately straight, the corresponding turning points are deleted; S234, filtering the turning points in the candidate turning point set of the central axis based on the periodic law of the meandering river bends: In the same curved section of a meandering river, there are usually only two inflection points. The polygonal set formed by the inflection point connection line and the central axis is defined as T = {t m |m=1,...,O}, at this time, the combination mode of adjacent river sections in T can be divided into two types: B-type mode and S-type mode, where: The B-type mode means that the adjacent polygons are located on one side of the central axis, and the S-type mode means that the adjacent polygons are located on both sides of the central axis. Only the inflection points of the S-type mode are retained, and the inflection points of the B-type mode are eliminated, where: Assume that the vertex set on the central axis of the meandering river obtained in S223 is represented as L = {p1, p2, ..., p m }, where P j (j∈1, 2, ..., m) is the jth vertex on the central axis of the meandering river, P k (k∈1, 2, ..., m) is the kth vertex on the central axis of the meandering river. The method for determining the combination mode of adjacent curved river sections is as follows: A, B, and C are any three adjacent inflection points on the central axis of the meandering river. From the inflection point A to the vertex P between the inflection points A and C, j Make vector Similarly, from the inflection point B to the vertex P between the inflection points B and C k Make vector make When the vector When the direction is opposite, the inflection point B is the S-mode inflection point. When the directions are the same, inflection point B is a B-mode inflection point. In this case, inflection point B is deleted, and then the next adjacent bend is determined until all inflection points are traversed.

11. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 10, characterized in that: The specific steps of step S231 are as follows: S2311, extracting all vertices on the meandering river central axis, taking the direction corresponding to the default storage order of the line segments in the meandering river central axis vector as the direction of the meandering river central axis, taking the starting point of the meandering river central axis as the tail of the first vector, and the point next to the starting point as the head of the first vector, traversing the vertices on the meandering river central axis to construct vectors in this way, until all vertices are traversed, the loop ends, and at this time, the vector group corresponding to the meandering river central axis is obtained; S2312. Calculate the change in the direction of the front and rear vectors in the vector group. When the change direction of the angle of the front and rear vectors changes, the point is considered to be an inflection point in a geometric sense.

12. The method for automatically extracting and quantitatively characterizing the characteristics of modern sedimentary structures in a single meander belt according to claim 1, characterized in that: The specific steps of step S3 are as follows: S31. Let the meander belt or a curved river section be represented by a polygon P = {v1, v2, ..., v n }, where v i (i∈1, 2, ..., n) is the i-th vertex on the boundary line, and its coordinates are (x i ,y i ); Since the polygon boundary is closed, v1 and v n The coordinates of the meandering river coincide with each other; the meandering river central axis is defined as L = {p1, p2, ..., p m }, where p j (j∈1, 2, ..., m) is the jth node on the central axis of the meandering river, and its coordinates are (x j ,y j ), the calculation method of basic geometric parameters is as follows: ①. Length of the central axis of the meandering river: ②, Straight line length: ③. Area: ④. Average width: ⑤. Curvature:

6. Circumference: S32. For point dams, their length and width are different from those of curved river sections: Let the two inflection points that make up the point dam be points v k , v l , whose coordinates are (x k ,y k ) and (x1, y1), the formula for calculating the point dam length is: ⑦. Length of point dam: The vertex set on the convex bank arc surrounding the point dam is set to P = {p1, p2, ..., p n }, then p i Point to line segment v k v l Draw a perpendicular line with foot at O ​​and point p i and O whose coordinates are (x i ,y i ) and (x o ,y o ), p i The distance between and O is Then the point dam width can be determined as: ⑧、Point dam width: w=MAX(||p i O||), MAX means taking the maximum value, i∈{1, 2,…, n}.

Images