TW202118993A - Method for exploring the underground silt - Google Patents

Abstract

A method for exploring the underground silt is provided to overcome the problem that the conventional underground silt exploring device cannot be used in a limited water area where the ships cannot travel freely. The method includes scanning the underground topography with a 3D sonar scanner to construct a cloud mode of the underground topography, separating at least one mud disposition area from the cloud mode of the underground topography, performing an algorithm for edge outline points on the mud disposition area(s) to divide the mud disposition area into a plurality of geometric triangular elements, calculating the volume of each geometric triangular element, and calculating the sum of the volumes of the plurality of geometric triangular elements to obtain a quantity of the mud in the at least one mud disposition area.

Description

Method for detecting silt volume in underwater terrain

The present invention relates to a measurement method, in particular to an underwater terrain silt volume detection method for measuring the silt volume of an underwater terrain environment.

Due to the complex terrain of the underwater environment and unpredictable operation risks, manual methods cannot be easily used to complete the measurement and operation of the amount of silt in the underwater terrain. Please refer to Figure 1, which is a conventional underwater silt detection device 9 which has a hull 91, a first probe 92, a second probe 93, and a global positioning device. A system 94 and a processing controller 95. The first probe 92 and the second probe 93 are arranged on the hull 91. The first probe 92 is used to emit high-frequency ultrasonic waves to measure the distance between the top layer of silt and the water surface. The two probes 93 are used to emit low-frequency ultrasonic waves to measure the distance between the bottom of the silt and the water surface. The global positioning system 94 is used to provide terrain longitude and latitude information. The processing controller 95 is connected to the first probe 92, the second probe 93 and the global Positioning system 94. The processing controller 95 realizes position recording and fixed-point measurement of the hull 91 through the global positioning system 94, and measures the hull position according to the measurement results of the first probe 92 and the second probe 93 The amount of silt in the underwater terrain. An embodiment similar to the conventional underwater silt detection device 9 has been disclosed in the Chinese Publication No. 204594437 "Underwater Silt Detection Device" patent case.

When the above-mentioned conventional underwater silt detection device 9 detects the amount of silt in a water area, since the first probe 92 and the second probe 93 are installed on the hull 91, the water area must be water surface capable. The unshielded water area for the hull to navigate freely can make the hull 91 move above the silt detection area to measure the amount of silt, so that the conventional underwater silt detection device 9 cannot be applied to a narrow restricted water area And a water surface to shield the waters, such as: the waters inside the pillars of the harbor terminal, the waters surrounding the marine working platform and the offshore wind turbine platform, etc.

In view of this, it is necessary to provide a method for detecting the amount of silt in underwater terrain to solve the above-mentioned problems.

In order to solve the above problems, the object of the present invention is to provide a method for detecting the amount of silt in the underwater terrain, which can measure the amount of silt in the underwater terrain environment in a sheltered water area.

The elements and components described in the full text of the present invention use the quantifiers "one" or "one" for convenience and to provide the general meaning of the scope of the present invention; in the present invention, it should be construed as including one or at least one, and single The concept of also includes the plural, unless it clearly implies other meanings.

The "point cloud" mentioned in the full text of the present invention refers to data obtained through a three-dimensional scanner. The scanned data is recorded in the form of points, and each point has a three-dimensional coordinate.

The "covered water area" mentioned in the full text of the present invention refers to the water area where the water surface cannot be used for the free navigation of the surveying vessel.

The underwater terrain silt volume detection method of the present invention includes: scanning the underwater terrain with a three-dimensional sonar scanner to construct an underwater terrain point cloud model; from the underwater terrain point cloud model, at least A silt deposition area is cut out; an edge contour point detection algorithm is performed on the silt deposition area to divide the silt deposition area into a number of triangular mesh geometric bodies; and the volume of each triangular mesh geometric body is calculated and totaled, In order to obtain a silt content of the silt deposition area.

Accordingly, the underwater terrain silt volume detection method of the present invention can scan the underwater terrain through the three-dimensional sonar scanner without being affected by water surface obstacles to construct the underwater terrain point cloud model. Perform the edge contour point detection algorithm for the silt deposition area where the silt content is to be calculated in the underwater topographic point cloud model to divide the silt deposition area into several triangle mesh geometric bodies, and then calculate each triangle mesh separately The volume of the geometry is added up. In this way, the underwater terrain silt volume detection method of the present invention has the effect of being applied to both sheltered waters and uncovered waters.

Wherein, the three-dimensional sonar scanner scans the underwater terrain from different azimuths to generate several fragment point cloud models, and the several fragment point cloud models execute an iterative nearest point algorithm to generate the several fragment point cloud models. The model is combined to form the underwater terrain point cloud model. In this way, the system can more accurately combine the several fragment point cloud models, and has the effect of improving the accuracy of establishing the underwater terrain point cloud model.

Among them, the edge contour point algorithm is Alpha Shape algorithm. In this way, it has the effect of improving the stability and adaptability of the operation.

Among them, before executing the edge contour point algorithm, the silt deposition area is divided into blocks through Delaunay's algorithm to form a triangulation network, and then the edge contour point algorithm is performed on the triangulation network to remove the The silt deposition area is divided into the triangle mesh geometry. In this way, it is possible to reduce the time complexity of executing the edge contour point algorithm, and it has the effect of improving the performance of the operation.

Wherein, the underwater topographic point cloud model is composed of a set of spatial points including several spatial points. After the silt deposition area is cut out, the spatial point compensation is performed on a horizontal reference surface of the silt deposition area. In this way, the system can add points to the sheltered bottom base level to establish a complete closed polygon of silt deposition, which has the effect of reducing the error between the calculated silt content and the actual silt content.

In order to make the above and other objectives, features and advantages of the present invention more comprehensible, the following describes the preferred embodiments of the present invention in conjunction with the accompanying drawings in detail as follows:

Please refer to Figure 2, which is a preferred embodiment of the underwater terrain silt volume detection method of the present invention, which includes a scanning step S1, a cutting step S2, a contour step S3, and a calculation step S4.

The scanning step S1 is to scan the underwater terrain with a three-dimensional sonar scanner (such as BlueView BV5000). The underwater terrain can be an unshielded water area or an underwater environment in a shielded water area to construct an underwater environment. Topographic point cloud (Point Cloud) model. In detail, in the scanning step S1, the three-dimensional sonar scanner is used to scan the underwater terrain from different directions to generate several fragment point cloud models; a one-point set matching algorithm is performed on the several fragment point cloud models In this embodiment, the point set matching algorithm is an Iterative Closest Point (ICP) algorithm to combine the several fragment point cloud models to form the underwater terrain point cloud model.

For example, when the scanning step S1 is used to detect the amount of silt in a sheltered water area (such as the water area inside the pillars of a harbor dock), the three-dimensional sonar scanner can be installed on a tripod and the three-dimensional sonar can be scanned The instrument scans the underwater environment of the sheltered water area to construct a three-dimensional point cloud model of the sheltered water area.

,                                                           　　（1） 其中，F(R,T)係為該第一片段點雲模型與該第二片段點雲模型之間的目標函數；R係為三維聲納掃描儀的旋轉；T係為三維聲納掃描儀的平移；Q_i 係為該第二片段點雲模型的空間點集合，P_i 係為該第一片段點雲模型的空間點集合，i=1,2,…,N，N為正整數。For example, the plurality of segment point cloud models have a first segment point cloud model and a second segment point cloud model, and P{P _i , i=1,2,...} represents the first segment point cloud The spatial point set of the model (Spatial Unorganized), with Q{Q _i , i=1,2,...} representing the spatial point set of the second segment point cloud model; the first segment point cloud model and the second segment The point cloud model is spatially matched, that is, rotation and translation are performed on the spatial point set of the second segment point cloud model, and an optimal solution formula is used to calculate the second segment point cloud model and the first segment point cloud model Perform spatial matching to minimize the objective function between the second segment point cloud model and the first segment point cloud model; repeat the above process for the remaining segment point cloud models of the plurality of segment point cloud models, by Therefore, the several segment point cloud models are combined to form the underwater terrain point cloud model. Among them, the best solution formula can be shown in the following formula (1):

, (1) Among them, F(R,T) is the objective function between the first segment point cloud model and the second segment point cloud model; R is the rotation of the three-dimensional sonar scanner; T is the three-dimensional sonar scanner translation; Q _i line segment for the second spatial point cloud model set point, P _i for space-based segment of the first set of points of the point cloud model, i = 1,2, ..., N , N Is a positive integer.

The cutting step S2 is used to cut out at least one silt deposition area from the underwater topographic point cloud model for subsequent steps to calculate the silt content in the silt deposition area. In this embodiment, the cutting step S2 is Random Sample Consensus (RANSAC) is used to find the normal direction of the seabed datum and the inclined plane of sedimentary silt from the underwater topographic point cloud model, and follow this normal The direction is estimated to be a subset of the point cloud of the seabed datum and the inclined surface of the sedimentation silt, so as to cut the silt deposition area from the point cloud model of the underwater topography.

,                                                                           　　（2） 其中，N係為該隨機抽樣一致性演算法的迭代次數；P係為該隨機抽樣一致性演算法執行一定次數後，從該水下地形點雲模型的空間點集合中選取到內群物件的機率；S為每次由該水下地形點雲模型的空間點集合中，隨機選取的空間點數量；e為該外群物件的機率。Specifically, the underwater terrain point cloud model system may be composed of a spatial point set containing several spatial points, and the random sampling consensus algorithm is composed of randomly selected spatial points from the spatial point set. An object of an inner group (Inliers); calculate and fit the inner group to establish a mathematical model; set a probability threshold of the mathematical model, and input the unselected spatial points in the set of spatial points into the mathematics one by one Model and perform calculations to determine whether the mathematical model meets the probability threshold. If it does, set the space point input to the mathematical model as the object of the inner group, and count the number of the object. If it does not meet the The spatial point input to the mathematical model is set as an Outliers object; the above process is repeated N times, and the inner group with the largest number of objects is used as the best model solution of the mathematical model, where the probability threshold The calculation formula of the value can be as shown in the following formula (2):

, (2) Among them, N is the number of iterations of the random sampling consensus algorithm; P is the random sampling consensus algorithm after a certain number of executions, selected from the set of spatial points of the underwater terrain point cloud model The probability of an object in the inner group; S is the number of space points randomly selected from the spatial point set of the underwater terrain point cloud model each time; e is the probability of the object in the outer group.

m，m=1,2,…,I，I為正整數；b

n，n=1,2,…,J，J為正整數；c係為該淤泥沉積區域的深度值。在本實施例中，該補點步驟S21係將該水平基準面的X軸向長度與Y軸向長度分別設定為25行及25列，藉此，對該水平基準面共計補上625個空間點。After the cutting step S2 is executed, that is, after the silt deposition area is cut out, a point compensation step S21 may be performed. The point compensation step S21 is used to perform spatial point compensation on a horizontal reference surface of the silt deposition area In this embodiment, the horizontal reference surface is the seabed reference surface of the underwater terrain. In detail, the point compensation step S21 can divide the X-axis length of the horizontal reference surface into m rows and divide the Y-axis length of the horizontal reference surface into n columns; set P(a, b, c ) Three-dimensional coordinates, where a

m, m=1,2,...,I, I is a positive integer; b

n, n=1,2,...,J, J is a positive integer; c is the depth value of the silt deposition area. In this embodiment, the point compensation step S21 is to set the X-axis length and Y-axis length of the horizontal reference plane to 25 rows and 25 columns, respectively, so that a total of 625 spaces are added to the horizontal reference plane. point.

The contour step S3 is used to execute an edge contour point detection algorithm for the silt deposition area. In this embodiment, the edge contour point detection algorithm is an Alpha Shape algorithm, which is used to divide the silt deposition area into numbers. A triangular mesh geometry for subsequent calculations to obtain the silt content of the silt deposition area.

Specifically, the Alpha Shape algorithm presets a detection circle; select two spatial points from the set of spatial points of the triangular mesh geometry, and determine whether the length of the line segment formed by the two spatial points is greater than the diameter of the detection circle , If the judgment result is yes, delete the triangle mesh geometry, re-select the other two space points from the set of space points of the triangle mesh geometry, and re-judge whether the length of the line segment formed by the other two space points is greater than the diameter ; If the judgment result is no, then further confirm whether the two-dimensional point passes through the detection circle, and any spatial point in the set of spatial points is not inside the detection circle, if the confirmation result is yes, then the two-dimensional point It is the edge contour point of the triangle mesh geometry, and the line segment formed by the two-space points is the edge contour line of the triangle mesh geometry; if the confirmation result is no, delete the triangle mesh geometry.

The calculation step S4 is used to calculate the volume of each triangle mesh geometry body and add them to obtain a silt content of the silt deposition area. Specifically, the calculation step S4 is to establish a Voronoi diagram based on the several triangular mesh geometry, and calculate the silt content of the silt deposition area based on the calculation method of the Voronoi diagram. The establishment of the Voronoi diagram and the calculation of the product model are common knowledge in the technical field of the present invention, and will not be repeated here.

In the method for detecting the amount of silt in underwater terrain of the present invention, in the contour step S3, before the edge contour point detection algorithm is executed, a subdivision step S5 may be performed on the silt deposition area. The subdivision step S5 is used for the silt The deposition area is divided into sections to form a triangulation network. In this embodiment, the silt deposition area is constructed to form the triangulation network through the Delaunay algorithm, and then the contour step S3 is used to form the triangulation network. The mesh executes the edge contour point detection algorithm to divide the silt deposition area into the plurality of triangular mesh geometric bodies.

具有二相關聯的德勞內三角形△abc，△abd時，分別求出該二德勞內三角形的外心r，s，該外心r所形成的外接圓會通過△abc的點a、點b及點c；該外心s所形成的外接圓會通過△abd的點a、點b及點d。若該Alpha Shape演算法所預設的檢測圓的圓心位於

上，且該檢測圓經過該點a、點b，則

即為由該Alpha Shape演算法所產生的三角網格幾何體的邊緣輪廓線。其中，可判斷該檢測圓的半徑是否大於該點a到

的最短距離，以及該檢測圓的半徑是否小於該點a到

的最遠距離，若判斷結果均相符，則表示該檢測圓係經過該點a、點b；若判斷結果不相符，則表示該檢測圓係未同時經過該點a、點b。Please refer to Figure 3. For example, if any line segment in the triangulation

When there are two associated Delaunay triangles △abc and △abd, find the outer center r, s of the second Delaunay triangle, and the circumcircle formed by the outer center r will pass through the points a and points of △abc b and point c; the circumcircle formed by the outer center s will pass through points a, b, and d of △abd. If the center of the detection circle preset by the Alpha Shape algorithm is at

Up, and the detection circle passes through the point a and point b, then

It is the edge contour line of the triangle mesh geometry generated by the Alpha Shape algorithm. Among them, it can be judged whether the radius of the detection circle is greater than the point a to

Is the shortest distance, and whether the radius of the detection circle is smaller than the point a to

If the judgment results are consistent, it means that the detection circle passes through the point a and point b; if the judgment result does not match, it means that the detection circle does not pass through the point a and point b at the same time.

只有一相關聯的德勞內三角形△abc時，求出該德勞內三角形的外心r，並以該外心r產生一射線

（Ray Casting），若該點a至該射線

的最短距離小於該檢測圓的半徑，則表示該

即為由該Alpha Shape演算法所產生的三角網格幾何體的邊緣輪廓線，否則，即表示該

不是該三角網格幾何體的邊緣輪廓線。Please refer to Figure 4, if any line segment in the triangulation

When there is only one associated Delaunay triangle △abc, find the outer center r of the Delaunay triangle, and generate a ray with the outer center r

(Ray Casting), if the point a to the ray

The shortest distance is less than the radius of the detection circle, it means the

It is the edge contour line of the triangle mesh geometry generated by the Alpha Shape algorithm. Otherwise, it means the

It is not the edge contour line of the triangle mesh geometry.

In summary, the underwater terrain silt volume detection method of the present invention can scan the underwater terrain through the three-dimensional sonar scanner without being affected by water surface obstacles to construct the underwater terrain point cloud Model, the silt deposition area in the underwater topographic point cloud model for which the silt content is to be calculated, execute the edge contour point detection algorithm to divide the silt deposition area into several triangle mesh geometric bodies, and then calculate each triangle separately The volume of the mesh geometry is added up. In this way, the underwater terrain silt volume detection method of the present invention has the effect of being applied to both sheltered waters and uncovered waters.

Although the present invention has been disclosed using the above-mentioned preferred embodiments, it is not intended to limit the present invention. Anyone who is familiar with the art without departing from the spirit and scope of the present invention may make various changes and modifications relative to the above-mentioned embodiments. The technical scope of the invention is protected. Therefore, the scope of protection of the invention shall be subject to the scope of the attached patent application.

﹝this invention﹞ S1: Scan step S2: Cutting step S21: Steps to make up points S3: contour step S4: Calculation steps S5: Splitting steps ﹝Used ﹞ 9: Underwater silt detection device 91: Hull 92: The first probe 93: second probe 94: Global Positioning System 95: processing controller

[Picture 1] A diagram of a conventional underwater terrain silt detection method. [Figure 2] A flowchart of a method according to a preferred embodiment of the present invention. [Figure 3] A triangulation diagram of the silt deposition area of a preferred embodiment of the present invention. [Figure 4] A triangulation diagram of the silt deposition area of a preferred embodiment of the present invention.

S1: Scan step

S2: Cutting step

S21: Steps to make up points

S3: contour step

S4: Calculation steps

S5: Splitting steps

Claims (5)

A method for detecting the amount of silt in underwater terrain, including: Scan the underwater terrain with a three-dimensional sonar scanner to construct an underwater terrain point cloud model; Cut out at least one silt deposition area from the underwater topographic point cloud model; Performing an edge contour point detection algorithm on the silt deposition area to divide the silt deposition area into a number of triangular mesh geometric bodies; and Calculate the volume of each triangle mesh geometry and add them to obtain a silt content in the silt deposition area. According to the method for detecting the amount of silt in the underwater terrain as described in item 1 of the scope of patent application, the three-dimensional sonar scanner scans the underwater terrain from different directions to generate several fragments of point cloud models, and the plurality of fragments The point cloud model executes an iterative nearest point algorithm to combine the several fragment point cloud models to form the underwater terrain point cloud model. The method for detecting the amount of silt in underwater terrain as described in item 1 of the scope of patent application, wherein the edge contour point algorithm is the Alpha Shape algorithm. The method for detecting the amount of silt in underwater terrain as described in item 3 of the scope of patent application, wherein, before executing the edge contour point algorithm, the silt deposition area is divided by Delaunay's algorithm to form a triangle Then, the edge contour point algorithm is executed on the triangular mesh to divide the silt deposition area into the plurality of triangular mesh geometric bodies. The underwater terrain silt volume detection method according to any one of items 1 to 4 of the scope of the patent application, wherein the underwater terrain point cloud model is composed of a collection of spatial points including a plurality of spatial points, and the silt After the deposition area is cut out, a horizontal datum of the silt deposition area is supplemented with spatial points.

Images