WO2018036138A1 - 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 - Google Patents
一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 Download PDFInfo
- Publication number
- WO2018036138A1 WO2018036138A1 PCT/CN2017/076569 CN2017076569W WO2018036138A1 WO 2018036138 A1 WO2018036138 A1 WO 2018036138A1 CN 2017076569 W CN2017076569 W CN 2017076569W WO 2018036138 A1 WO2018036138 A1 WO 2018036138A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- point cloud
- cloud data
- thin
- dimensional shape
- dimensional
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Definitions
- the invention relates to the technical field of designing a bearing component of an aerospace structure, in particular to a method for processing a three-dimensional shape point cloud data of a thin-walled shell obtained by digital photography.
- the launch vehicle is the premise and basis for accelerating China's becoming a world space power and realizing the country's peaceful development and utilization of space.
- China is demonstrating the development of a heavy-duty launch vehicle with a core diameter of 8.5 meters and a carrying capacity of 100 tons.
- structural weight reduction is a necessary and effective way. Since the propellant tank accounts for 80% of the volume of the launch vehicle, taking the new generation launch vehicle CZ-5 (5 meters in diameter) as an example, the weight of the tank structure accounts for 25% of the total weight of the rocket structure before the propellant is added.
- the lightweight design of the tank is important for improving the carrying capacity of the rocket.
- the propellant tank structure of the launch vehicle is composed of a front bottom, a cylinder section and a rear bottom.
- the cylinder section is composed of a plurality of wall plates welded, and the weight of the wall plate directly affects the weight of the box body.
- the tank wall which is mainly subjected to the axial pressure is often designed as a mesh reinforced structure.
- most of the propellant tanks of large and heavy-duty launch vehicles at home and abroad use lightweight aluminum alloy or aluminum-lithium alloy.
- the weight of the tank siding in the form of grid reinforcement accounts for 10.5% of the total weight of the rocket before the propellant is added.
- the ratio will be greatly improved.
- the lightweight design theory and method of grid-reinforced siding is an unavoidable requirement for ensuring the carrying capacity of heavy rockets.
- the large carrying capacity of the heavy-duty launch vehicle makes the grid-reinforced tank slab must have the mechanical properties of high axial bearing capacity, and the increase of the thickness-to-thickness ratio of the heavy-duty launch vehicle grid-reinforced tank siding, the initial geometry of the siding The sensitivity of the defect is also increased, which makes the actual carrying capacity of the grid-reinforced tank much smaller than the perfect predicted value. Therefore, the development of an accurate prediction method for the ultimate bearing capacity of the grid-reinforced shell structure is to carry out the lightweight design of the grid-reinforced tank. Key technology.
- the introduction of the measured initial geometric defect into the prediction model is the key technology for developing an accurate prediction method for the axial bearing capacity of the grid-reinforced cylindrical shell structure.
- the measurement accuracy is continuously improved, and the three-dimensional shape point cloud data of the grid-reinforced column-shell structure becomes very large.
- the three-dimensional topographical point cloud data of a grid-reinforced cylinder with a diameter of 4.5 meters and a height of 2.2 meters can reach millions of orders.
- the bounding box method is a three-dimensional shape point cloud data reduction method commonly used in practical engineering.
- the size of the bounding box is arbitrarily specified by the user.
- the accuracy between the constructed model and the original three-dimensional shape point cloud data cannot be guaranteed, and the point cloud feature is easily lost. It is not a high-fidelity three-dimensional shape point cloud data reduction method.
- the curvature sampling method retains a small number of data points in the small curvature region, and retains enough data points in the large curvature region, thereby achieving accurate and complete representation of the surface features with high precision, but the curvature calculation consumes a large amount.
- the computer resources, the algorithm is less efficient and less efficient.
- the three-dimensional shape point cloud data processing method for the thin-walled shell obtained by digital camera Under the premise of ensuring the three-dimensional shape feature, the three-dimensional shape point cloud data is efficiently reduced, and the grid reinforcement is further improved.
- the cylindrical shell structure is analyzed to predict the efficiency.
- the invention mainly aims at the multi-dimensional three-dimensional shape point cloud data measured by the thin-walled shell obtained by the existing digital camera, and proposes a method for processing the three-dimensional shape point cloud data of the thin-walled shell obtained by digital photography. Coordinate transformation of wall three-dimensional shape point cloud data, filtering processing of three-dimensional topographic point cloud data of thin-walled shell And the streamlined processing of the three-dimensional shape point cloud data of the thin-walled shell, and the three-dimensional topographic point cloud data containing the initial geometric defect features of the grid-reinforced cylindrical shell is simplified, and the mesh-reinforced column-shell structure is efficiently adjusted and accurately predicted. .
- a method for processing a three-dimensional shape point cloud data of a thin-walled shell obtained by digital camera which comprises the following steps:
- the first step is the coordinate transformation of the three-dimensional shape point cloud data of thin-walled shells.
- the complete three-dimensional shape point cloud data of the thin-walled shell is obtained by multiple measurements, and the coordinate transformation method is used to coordinate the three-dimensional topographic point cloud data; the complete three-dimensional shape The point cloud data is measured data divided into a plurality of different coordinate systems; the coordinate transformation method includes rotation, translation, and scale change processing of the three-dimensional top point cloud data.
- the second step is the filtering treatment of the three-dimensional shape point cloud data of the thin-walled shell
- the filtering algorithm for the three-dimensional shape point cloud data of the thin-walled shell is as follows:
- the filter processing algorithm is to set a threshold value. When the difference between the three-dimensional shape point cloud data and its neighbor mean exceeds the threshold, it is equal to the average of the neighborhood, and vice versa. .
- the mathematical representation of the filter processing algorithm is:
- p i is the filtered value of the measurement point i data; It is the neighborhood average of the measuring point i; q i is the original radius of the measuring point i; M is the threshold value; N is the number of measuring points; v ij is the neighboring mean weight coefficient of the measuring point j to the measuring point i; It is the neighborhood radius of the measuring point i; A i is the three-dimensional coordinate vector of the measuring point i.
- the third step is the streamlining of the three-dimensional shape point cloud data of thin-walled shells.
- the first step of the second-order filtering of the point cloud data after the filtering process is performed, so as to reduce the size of the point cloud data without losing the three-dimensional features.
- the first reduction processing method is a bounding box method, a random adoption method, a uniform grid method or a non-uniform grid method.
- step 3.1 Perform the local surface fitting of the scattered three-dimensional shape point cloud data processed in step 3.1), solve the fitted surface equation, obtain the curvature value of the three-dimensional shape point cloud data, and further solve the three-dimensional shape point cloud data neighbor.
- the local surface fitting method is a circle fitting or a paraboloid fitting or the like.
- Thin-walled shell three-dimensional shape point cloud data reduction process is based on the principle that the point cloud data curvature is smaller than the field curvature mean value, and the point cloud data is streamlined. If the curvature value of the three-dimensional shape point cloud data is smaller than the mean value of the neighborhood curvature, the three-dimensional topographic point cloud data is reduced, and the three-dimensional topographic point cloud data is efficiently and streamlined; otherwise, the three-dimensional topographic point cloud data is retained.
- the invention has the beneficial effects that the present invention provides a method for processing a three-dimensional shape point cloud data of a thin-walled shell obtained by digital photography, and is mainly directed to a three-dimensional three-dimensional three-dimensional measurement obtained by a conventional digital camera.
- the shape point cloud data through the coordinate transformation of the thin-walled three-dimensional shape point cloud data, the filtering processing of the thin-walled shell three-dimensional shape point cloud data, and the streamlining processing of the thin-walled shell three-dimensional shape point cloud data, streamlining the grid plus Three-dimensional topographic point cloud data of the initial geometric defect features of the ribbed shell.
- the simplification method based on non-uniform grid method and curvature sampling makes the point cloud simplification data retain the main features of the original data (ie, the initial geometric defect features of the thin-walled shell structure), and improves the curvature solution in the simplification process.
- the efficiency which realizes the efficient processing of the measured three-dimensional shape point cloud data of thin-walled shells, further improves the analysis and prediction efficiency of the grid-reinforced column shell structure, and is very promising to become the aerospace field of launch vehicle and missile design in China.
- FIG. 1 is a flowchart of implementing a method for processing a three-dimensional shape point cloud data of a thin-walled shell obtained by digital photography according to an embodiment of the present invention
- Figure 2 (a) is a three-dimensional topographic point cloud data map before processing
- Figure 2 (b) is a three-dimensional topographic point cloud data map after processing
- Figure 3 is a graph showing the prediction results of the ultimate bearing capacity of thin-walled shell structures based on point cloud data of different three-dimensional shapes.
- FIG. 1 is a flowchart of implementing a method for processing a three-dimensional shape point cloud data of a thin-walled shell obtained by digital photography according to an embodiment of the present invention.
- a method for processing a three-dimensional spatial coordinate point cloud data of a thin-walled shell obtained by digital photography according to an embodiment of the present invention includes:
- the first step is the coordinate transformation of the three-dimensional shape point cloud data of thin-walled shells.
- the thin-walled shell structure with a diameter of 1.0 m, a height of 0.6 m and a wall thickness of 2.0 mm was measured by digital photography method, and the three-dimensional shape point cloud data of six different regions of the thin-walled shell were obtained.
- the three-dimensional shape point cloud data is transformed by the rotation, translation and proportional change processing of the three-dimensional shape point cloud data, and the overall three-dimensional shape point cloud data of the thin-walled shell structure is obtained.
- the second step is the filtering treatment of the three-dimensional shape point cloud data of the thin-walled shell
- the overall three-dimensional shape of the thin-walled shell structure obtained in the first step The cloud data is filtered to remove noise;
- the third step is to simplify the three-dimensional shape point cloud data of thin-walled shells.
- the 3D topographic point cloud data after the second step of filtering is first streamlined by the non-uniform grid method. At this time, the 3D topography point cloud data is reduced from 1 million to 600,000.
- Figure 2 (a) is a three-dimensional topographic point cloud data map before processing, in which x, y, z are spatial coordinates;
- Figure 2 (b) is a processed three-dimensional topographic point cloud data map; in order to further characterize the present invention
- Figure 3 shows the prediction results of ultimate bearing capacity of thin-walled shell structures based on point cloud data of different three-dimensional shapes. The prediction result of the point cloud data modification is too high, and does not reflect the initial geometric defects of the thin-walled shell.
- the multi-dimensional three-dimensional shape point cloud data repairing result reflects the structural topography, but its prediction model repair The consumption time is up to ten hours, and the efficiency is extremely low; the result of the traditional streamlined point cloud data repair and the multi-dimensional three-dimensional shape point cloud data are quite different, because it loses some geometric defect features;
- the three-dimensional topography point cloud data trimming result of the present invention is similar to the one-dimensional three-dimensional topography point cloud data trimming result, indicating the accuracy of the present invention, and the predictive model trimming duration of the present invention is 30 minutes, and the efficiency is obtained. Significantly improved.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Processing Or Creating Images (AREA)
Abstract
一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，涉及航空航天结构主承力构件设计领域，该方法包括以下步骤：1)在薄壁壳三维形貌测量过程中，通过多次测量获取薄壁壳的完整的三维形貌点云数据，采用坐标变换方法对三维形貌点云数据进行坐标变换；2)将第一步坐标变换后得到的三维形貌点云数据采用滤波方法进行滤波处理；3)薄壁壳三维形貌点云数据的精简处理。该方法的有益效果为：针对数字摄影技术特点，保留薄壁壳结构几何缺陷特征，提高曲率求解效率，实现薄壁壳实测三维空间坐标点云数据的高效处理，进一步提高网格加筋柱壳结构分析预测效率，有望成为航空航天领域中三维形貌测量数据处理的关键技术之一。
Description
本发明涉及航空航天结构主承力构件设计技术领域，尤其涉及一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法。
运载火箭是加快推动中国成为世界航天强国，实现国家和平开发利用空间的前提和基础。为实施载人登月计划、完成深空探测任务，我国正在论证研制芯级直径8.5米、运载能力百吨级的重型运载火箭。为保证火箭的运载能力，结构减重是必须且有效的途径。由于推进剂贮箱占运载火箭体积的80％，以新一代运载火箭CZ-5(直径5米)为例，贮箱结构重量占未加注推进剂前火箭结构总重的25％，为此贮箱的结构轻量化设计对提升火箭运载能力十分重要。运载火箭的推进剂贮箱结构由前底、筒段、后底构成，其中筒段由若干壁板焊接组成，壁板的重量直接影响箱体的重量。为提高壁板的强度、减轻壁板的重量，承受轴压为主的贮箱壁板往往设计成网格加筋结构。考虑到材料稳定性、结构承载效率以及制造成本等综合因素，国内外大型和重型运载火箭的推进剂贮箱大多采用轻质铝合金或铝锂合金。以新一代运载火箭CZ-5为例，其以网格加筋形式的贮箱壁板重量，占火箭未加注推进剂前总重的10.5％，据重型运载火箭(直径8.5米)的前期论证，该比例还会有较大提高，为此网格加筋壁板的轻量化设计理论和方法是保证重型火箭运载能力不可回避的需求。重型运载火箭的大运载能力使得网格加筋贮箱壁板必须具备高轴压承载力的力学性能，加之重型运载火箭网格加筋贮箱壁板的径厚比提高，壁板对初始几何缺陷的敏感性也随之提高，这使得网格加筋贮箱的实际承载力远小于完美预测值。因此，发展准确预测网格加筋筒壳结构轴压极限承载力的方法是开展网格加筋贮箱轻量化设计的
关键技术。
考虑到初始几何缺陷是影响网格加筋柱壳结构轴压承载力的重要因素，将实测的初始几何缺陷引入预测模型是发展准确预测网格加筋柱壳结构轴压承载力方法的关键技术。随着三维形貌测量方法的发展，特别是数字摄像方法，测量精度不断提高，网格加筋柱壳结构的三维形貌点云数据变得十分庞大。通常来说，直径4.5米高度2.2米的网格加筋筒壳的三维形貌点云数据可达到百万量级。百万量级点云数据不仅在存储、处理和显示过程中需要消耗大量的时间和计算机资源，而且降低网格加筋柱壳模型的修调效率，影响模型的光顺性。因此，实测三维形貌点云数据的精简是考虑实测几何缺陷网格加筋柱壳结构预测分析的关键方法。
在目前的逆向工程中，包围盒法是实际工程中常用的三维形貌点云数据精简方法。但包围盒的大小是由用户任意规定的，无法保证所构建模型与原始三维形貌点云数据之间的精度，易丢失点云特征，并不是高保真的三维形貌点云数据精简方法。相比于包围盒法，曲率采样方法在小曲率区域保留少量数据点，在大曲率区域保留足够多的数据点，从而实现精确完整地表示曲面特征，具有较高的精度，但曲率计算消耗大量的计算机资源，该算法精简效率较低。因此，亟需开展面向数字摄像获得的薄壁壳实测的三维形貌点云数据处理方法研究，在保证三维形貌特征的前提下，高效精简三维形貌点云数据，进一步提高网格加筋柱壳结构分析预测效率。
发明内容
本发明主要针对现有数字摄像获得的薄壁壳实测的百万量级的三维形貌点云数据，提出一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，通过薄壁三维形貌点云数据的坐标变换、薄壁壳三维形貌点云数据的滤波处理
以及薄壁壳三维形貌点云数据的精简处理，精简含有网格加筋柱壳初始几何缺陷特征的三维形貌点云数据，达到网格加筋柱壳结构高效修调、准确预测的目的。
为了达到上述目的，本发明的技术方案为：
一种面向数字摄像获得的薄壁壳实测三维形貌点云数据处理方法，具体包括以下步骤：
第一步，薄壁壳三维形貌点云数据的坐标变换
在薄壁壳三维形貌测量过程中，通过多次测量获取薄壁壳的完整的三维形貌点云数据，采用坐标变换方法对三维形貌点云数据进行坐标变换；所述的完整三维形貌点云数据为被分成多块不同坐标系下的测量数据；所述的坐标变换方法包括三维形貌点云数据的旋转、平移以及比例变化处理。
第二步，薄壁壳三维形貌点云数据的滤波处理
测量数据中噪声不仅直接影响测量的质量，还增加后续处理工作的难度，因此需要将第一步坐标变换后得到的三维形貌点云数据采用滤波方法进行滤波处理用于消除噪声，所述的薄壁壳三维形貌点云数据的滤波处理算法具体为：
由于噪声一般和邻域样本数据值相差较大，当插值超过一定门限时才被认为是噪声。因此，滤波处理算法的基本思想是设定门限值，当三维形貌点云数据与其邻域平均值的差值超过门限时，令其等于该邻域的平均值，反之，数据值不变。滤波处理算法的数学表示为：
其中pi为测点i数据滤波后的值；
为测点i的邻域平均值；qi为测点i的原始半径；M为门限值；N为测点数目；vij为测点j对测点i邻域平均值权系数；L为测点i邻域半径；Ai为测点i的三维坐标向量。
第三步，薄壁壳三维形貌点云数据的精简处理
3.1)将第二步滤波处理后的三维形貌点云数据进行第一次精简处理，目的是在不丢失三维形貌特征的前提下，降低点云数据规模。所述的第一次精简处理方法为包围盒法、随机采用法、均匀网格法或非均匀网格法等。
3.2)将步骤3.1)处理后的散乱的三维形貌点云数据进行局部曲面拟合，求解拟合的曲面方程，获得三维形貌点云数据的曲率值，进一步求解三维形貌点云数据邻域的曲率均值。所述的局部曲面拟合方法为圆拟合或抛物面拟合等。
3.3)薄壁壳三维形貌点云数据精简过程基于点云数据曲率小于其领域曲率均值原则，精简点云数据。若三维形貌点云数据的曲率值小于邻域曲率均值，则精简去除该三维形貌点云数据，实现三维形貌点云数据的高效精简；反之，保留三维形貌点云数据。
本发明的有益效果为：本发明提供的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，主要针对现有数字摄像获得的薄壁壳实测的百万量级的三维形貌点云数据，通过薄壁三维形貌点云数据的坐标变换、薄壁壳三维形貌点云数据的滤波处理以及薄壁壳三维形貌点云数据的精简处理，精简含有网格加筋柱壳初始几何缺陷特征的三维形貌点云数据。其中，基于非均匀网格法与曲率采样的精简方法，使得点云精简数据保留了原始数据的主要特征(即薄壁壳结构的初始几何缺陷特征)的同时，提高了精简过程中的曲率求解效率，从而实现了薄壁壳实测三维形貌点云数据的高效处理，进一步提高网格加筋柱壳结构分析预测效率，十分有望成为我国运载火箭、导弹设计等航空航天领域
中三维形貌测量数据处理的关键方法之一。
图1为本发明实施例提供的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法的实现流程图；
图2(a)为处理前的三维形貌点云数据图；
图2(b)为处理后的三维形貌点云数据图；
图3为基于不同三维形貌点云数据的薄壁壳结构极限承载力预测结果图。
为使本发明解决的方法问题、采用的方法方案和达到的方法效果更加清楚，下面结合附图和实施例对本发明作进一步的详细说明。可以理解的是，此处所描述的具体实施例仅仅用于解释本发明，而非对本发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与本发明相关的部分而非全部内容。
图1为本发明实施例提供的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法的实现流程图。如图1所示，本发明实施例提供的一种面向数字摄影获得的薄壁壳实测三维空间坐标点云数据处理方法包括：
第一步，薄壁壳三维形貌点云数据的坐标变换
在薄壁壳三维形貌测量过程中，采用数字摄影方法，测量直径1.0米、高度0.6米、壁厚2.0毫米的薄壁壳结构，获得薄壁壳6个不同区域的三维形貌点云数据；采用三维形貌点云数据的旋转、平移以及比例变化处理，对三维形貌点云数据进行坐标变换，得到薄壁壳结构的整体的三维形貌点云数据。
第二步，薄壁壳三维形貌点云数据的滤波处理
采用超限邻域平均滤波算法，对第一步得到的薄壁壳结构的整体的三维形
貌点云数据进行滤波处理，消除噪声；
第三步，薄壁壳三维形貌点云数据的精简方法
3.1)将第二步滤波处理后的三维形貌点云数据采用非均匀网格法进行第一次精简，此时，三维形貌点云数据由100万降至60万。
3.2)对3.1)得到的三维形貌点云数据，采用抛物面拟合局部曲面，求解拟合的曲面方程，获得三维形貌点云数据的曲率值，进一步，求解三维形貌点云数据邻域(20毫米)的曲率均值。
3.3)对比分析三维形貌点云数据的曲率值与邻域的曲率均值，若三维形貌点云数据的曲率值小于邻域曲率均值，则精简去除该三维形貌点云数据，反之，保留三维形貌点云数据，循环处理三维形貌点云数据中的每个数据，完成薄壁壳三维形貌点云数据的精简，此时，三维形貌点云数据降至1万。
图2(a)为处理前的三维形貌点云数据图,图中x、y、z为空间坐标；图2(b)为处理后的三维形貌点云数据图；为了进一步表征本发明具有高效准确的特点，图3给出了基于不同三维形貌点云数据的薄壁壳结构极限承载力预测结果。无点云数据修调的预测结果偏高，没有体现薄壁壳的初始几何缺陷；基于百万量级的三维形貌点云数据修调结果体现了结构的形貌特征，但其预测模型修调耗时长达十几小时，效率极低；基于传统的精简点云数据修调结果与百万量级的三维形貌点云数据结果相差较大，原因是其丢失了部分几何缺陷特征；基于本发明的三维形貌点云数据修调结果与百万量级的三维形貌点云数据修调结果相近，说明本发明的准确性，且本发明预测模型修调时长为30分钟，效率得到显著提升。
Claims (5)
- 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，其特征在于以下步骤：第一步，薄壁壳三维形貌点云数据的坐标变换在薄壁壳三维形貌测量过程中，通过多次测量获取薄壁壳的完整的三维形貌点云数据，采用坐标变换方法对三维形貌点云数据进行坐标变换；第二步，薄壁壳三维形貌点云数据的滤波处理将第一步坐标变换后得到的三维形貌点云数据采用滤波方法进行滤波处理，消除噪声；所述的薄壁壳三维形貌点云数据的滤波处理算法为：设置门限，当三维形貌点云数据与其邻域平均值的差值超过门限时，令其等于该邻域的平均值，反之，数据值不变；所述的薄壁壳三维形貌点云数据的滤波处理算法的数学表示为：
其中，pi为测点i数据滤波后的值；为测点i的邻域平均值；qi为测点i的原始半径；M为门限；N为测点数目；vij为测点j对测点i邻域平均值权系数；L为测点i邻域半径；Ai为测点i的三维坐标向量；
第三步，薄壁壳三维形貌点云数据的精简处理3.1)将第二步滤波处理后的三维形貌点云数据进行第一次精简处理，在不丢失三维形貌特征的前提下，降低点云数据规模；3.2)将步骤3.1)处理后的三维形貌点云数据进行局部曲面拟合，求解拟合的曲面方程，获得三维形貌点云数据的曲率值，进一步求解三维形貌点云数 据邻域的曲率均值；3.3)若三维形貌点云数据的曲率值小于邻域曲率均值，则精简去除该三维形貌点云数据，实现三维形貌点云数据的高效精简；反之，保留三维形貌点云数据。 - 根据权利要求1所述的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，其特征在于，第一步中所述的坐标变换方法包括三维形貌点云数据的旋转、平移以及比例变化处理。
- 根据权利要求1或2所述的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，其特征在于，步骤3.1)中所述的第一次精简处理方法为包围盒法、随机采用法、均匀网格法或非均匀网格法。
- 根据权利要求1或2所述的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，其特征在于，步骤3.2)中所述的局部曲面拟合方法为圆拟合或抛物面拟合。
- 根据权利要求3所述的一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法，其特征在于，步骤3.2)中所述的局部曲面拟合方法为圆拟合或抛物面拟合。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018549816A JP6587305B2 (ja) | 2016-08-24 | 2017-03-14 | デジタル撮影によって得られる薄壁シェルの実測された３ｄプロファイルポイントクラウドデータ処理方法 |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610714898.8A CN106372283B (zh) | 2016-08-24 | 2016-08-24 | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 |
| CN201610714898.8 | 2016-08-24 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2018036138A1 true WO2018036138A1 (zh) | 2018-03-01 |
Family
ID=57878168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2017/076569 WO2018036138A1 (zh) | 2016-08-24 | 2017-03-14 | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 |
Country Status (3)
| Country | Link |
|---|---|
| JP (1) | JP6587305B2 (zh) |
| CN (1) | CN106372283B (zh) |
| WO (1) | WO2018036138A1 (zh) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685848A (zh) * | 2018-12-14 | 2019-04-26 | 上海交通大学 | 一种三维点云与三维传感器的神经网络坐标变换方法 |
| CN109816788A (zh) * | 2019-01-17 | 2019-05-28 | 中国公路工程咨询集团有限公司 | 一种基于三维激光点云数据的三维地质体建模方法 |
| CN110992274A (zh) * | 2019-11-12 | 2020-04-10 | 中国科学院自动化研究所 | 复杂曲面点云强噪音去除方法及系统 |
| CN111311576A (zh) * | 2020-02-14 | 2020-06-19 | 易思维（杭州）科技有限公司 | 基于点云信息的缺陷检测方法 |
| CN111369610A (zh) * | 2020-03-05 | 2020-07-03 | 山东交通学院 | 基于可信度信息的点云数据粗差定位和剔除方法 |
| CN111754421A (zh) * | 2020-06-01 | 2020-10-09 | 西安工业大学 | 改进的导向滤波三维散乱点云快速光顺方法 |
| CN112927287A (zh) * | 2021-03-24 | 2021-06-08 | 和数科技(浙江)有限公司 | 一种目标物体的表型数据分析方法、存储介质及终端 |
| CN113742850A (zh) * | 2021-09-09 | 2021-12-03 | 北京航空航天大学 | 基于速度损失的固液动力飞行器运载能力预估方法和系统 |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372283B (zh) * | 2016-08-24 | 2018-06-08 | 大连理工大学 | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 |
| CN109408939B (zh) * | 2018-10-18 | 2022-11-29 | 燕山大学 | 一种兼顾应力和位移约束的薄板结构加强筋分布优化的改进方法 |
| CN110096772B (zh) * | 2019-04-17 | 2022-10-25 | 大连理工大学 | 一种面向航空航天薄壳结构的形位偏差特征库建立方法 |
| US20210056243A1 (en) * | 2019-04-17 | 2021-02-25 | Dalian University Of Technology | Method for establishing the geometrical imperfection database for the aerospace thin-walled structure |
| CN110530306B (zh) * | 2019-08-27 | 2020-10-20 | 大连理工大学 | 一种基于实测跳动数据的典型回转体零件表征方法 |
| CN113218330B (zh) * | 2021-04-27 | 2022-04-08 | 大连理工大学 | 基于pcl库的大型薄壁构件测量及数据处理方法 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750730A (zh) * | 2012-06-15 | 2012-10-24 | 北京理工大学 | 一种特征保持的点云数据精简方法 |
| CN103310481A (zh) * | 2013-05-20 | 2013-09-18 | 东南大学 | 一种基于模糊熵迭代的点云精简方法 |
| US20130332110A1 (en) * | 2012-06-08 | 2013-12-12 | Massachusetts Institute Of Technology | Non-iterative mapping of capped cylindrical environments |
| CN103886555A (zh) * | 2014-03-12 | 2014-06-25 | 北京昊峰东方科技有限公司 | 一种基于海量三维激光扫描点云数据的处理方法 |
| CN106372283A (zh) * | 2016-08-24 | 2017-02-01 | 大连理工大学 | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11232489A (ja) * | 1998-02-18 | 1999-08-27 | Ricoh Co Ltd | ３次元ポリゴンメッシュのデータ削減方法及び装置 |
| US8355601B2 (en) * | 2010-01-15 | 2013-01-15 | Seiko Epson Corporation | Real-time geometry aware projection and fast re-calibration |
| US9767598B2 (en) * | 2012-05-31 | 2017-09-19 | Microsoft Technology Licensing, Llc | Smoothing and robust normal estimation for 3D point clouds |
| JP2014145735A (ja) * | 2013-01-30 | 2014-08-14 | Nikon Corp | 形状測定装置、構造物製造システム、評価装置、形状測定方法、構造物製造方法、及び形状測定プログラム |
| CN104573144A (zh) * | 2013-10-14 | 2015-04-29 | 鸿富锦精密工业（深圳）有限公司 | 量测设备离线点云模拟系统及方法 |
-
2016
- 2016-08-24 CN CN201610714898.8A patent/CN106372283B/zh active Active
-
2017
- 2017-03-14 JP JP2018549816A patent/JP6587305B2/ja active Active
- 2017-03-14 WO PCT/CN2017/076569 patent/WO2018036138A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130332110A1 (en) * | 2012-06-08 | 2013-12-12 | Massachusetts Institute Of Technology | Non-iterative mapping of capped cylindrical environments |
| CN102750730A (zh) * | 2012-06-15 | 2012-10-24 | 北京理工大学 | 一种特征保持的点云数据精简方法 |
| CN103310481A (zh) * | 2013-05-20 | 2013-09-18 | 东南大学 | 一种基于模糊熵迭代的点云精简方法 |
| CN103886555A (zh) * | 2014-03-12 | 2014-06-25 | 北京昊峰东方科技有限公司 | 一种基于海量三维激光扫描点云数据的处理方法 |
| CN106372283A (zh) * | 2016-08-24 | 2017-02-01 | 大连理工大学 | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685848A (zh) * | 2018-12-14 | 2019-04-26 | 上海交通大学 | 一种三维点云与三维传感器的神经网络坐标变换方法 |
| CN109685848B (zh) * | 2018-12-14 | 2023-06-09 | 上海交通大学 | 一种三维点云与三维传感器的神经网络坐标变换方法 |
| CN109816788A (zh) * | 2019-01-17 | 2019-05-28 | 中国公路工程咨询集团有限公司 | 一种基于三维激光点云数据的三维地质体建模方法 |
| CN110992274B (zh) * | 2019-11-12 | 2023-06-02 | 中国科学院自动化研究所 | 复杂曲面点云强噪音去除方法及系统 |
| CN110992274A (zh) * | 2019-11-12 | 2020-04-10 | 中国科学院自动化研究所 | 复杂曲面点云强噪音去除方法及系统 |
| CN111311576A (zh) * | 2020-02-14 | 2020-06-19 | 易思维（杭州）科技有限公司 | 基于点云信息的缺陷检测方法 |
| CN111311576B (zh) * | 2020-02-14 | 2023-06-02 | 易思维（杭州）科技有限公司 | 基于点云信息的缺陷检测方法 |
| CN111369610A (zh) * | 2020-03-05 | 2020-07-03 | 山东交通学院 | 基于可信度信息的点云数据粗差定位和剔除方法 |
| CN111369610B (zh) * | 2020-03-05 | 2022-09-06 | 山东交通学院 | 基于可信度信息的点云数据粗差定位和剔除方法 |
| CN111754421A (zh) * | 2020-06-01 | 2020-10-09 | 西安工业大学 | 改进的导向滤波三维散乱点云快速光顺方法 |
| CN111754421B (zh) * | 2020-06-01 | 2023-08-18 | 西安工业大学 | 改进的导向滤波三维散乱点云快速光顺方法 |
| CN112927287A (zh) * | 2021-03-24 | 2021-06-08 | 和数科技(浙江)有限公司 | 一种目标物体的表型数据分析方法、存储介质及终端 |
| CN113742850B (zh) * | 2021-09-09 | 2023-05-16 | 北京航空航天大学 | 基于速度损失的固液动力飞行器运载能力预估方法和系统 |
| CN113742850A (zh) * | 2021-09-09 | 2021-12-03 | 北京航空航天大学 | 基于速度损失的固液动力飞行器运载能力预估方法和系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP6587305B2 (ja) | 2019-10-09 |
| JP2019510975A (ja) | 2019-04-18 |
| CN106372283A (zh) | 2017-02-01 |
| CN106372283B (zh) | 2018-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018036138A1 (zh) | 一种面向数字摄影获得的薄壁壳实测三维形貌点云数据处理方法 | |
| CN104866673B (zh) | 一种轴压加筋柱壳的开口补强方法 | |
| Upadhyay et al. | Numerical analysis perspective in structural shape optimization: A review post 2000 | |
| WO2014161364A1 (zh) | 一种轴压筒壳结构承载力折减因子确定方法 | |
| WO2020211007A1 (zh) | 一种面向航空航天薄壳结构的形位偏差特征库建立方法 | |
| CN109658431B (zh) | 基于区域生长的岩体点云平面提取方法 | |
| CN112036058A (zh) | 一种非完善薄壁结构的快速缺陷敏感性分析方法 | |
| CN106548484A (zh) | 基于二维凸包的产品模型散乱点云边界特征提取方法 | |
| CN106372282B (zh) | 一种体现制造几何缺陷的三维有限元模型修调方法 | |
| CN108491654B (zh) | 一种三维实体结构拓扑优化方法及系统 | |
| Peng et al. | Automatic 3D image based finite element modelling for metallic foams and accuracy verification of digital volume correlation | |
| Kozma et al. | Computer tomography based reconstruction of metal matrix syntactic foams | |
| CN113672849A (zh) | 一种基于轴向通量展开的一步法输运计算方法及系统 | |
| CN110555189B (zh) | 一种基于反向计算思维的空间插值方法 | |
| CN108875260B (zh) | 一种热喷涂涂层性能参数计算方法 | |
| CN116701969A (zh) | 一种用于气膜孔点云的处理及轴线估计方法 | |
| CN108731616B (zh) | 一种基于云模型的螺旋锥齿轮齿面测量点自适应分布方法 | |
| CN115564820A (zh) | 基于贪婪投影三角化的体积确定方法、系统、设备及介质 | |
| CN110096772A (zh) | 一种面向航空航天薄壳结构的形位偏差特征库建立方法 | |
| CN115408796A (zh) | 一种多孔结构建模方法、装置、设备及应用 | |
| CN104458495B (zh) | 颗粒增强金属基复合材料增强体颗粒体积分数的测量方法 | |
| CN112131665B (zh) | 一种基于网格分类的火箭流固载荷混合转换方法 | |
| CN114186297A (zh) | 一种基于拓扑优化的变密度随形点阵结构设计方法 | |
| Zhu et al. | Influence of initial imperfections on PMMA spherical shell | |
| Zhao et al. | A preprocessing method of 3D point clouds registration in urban environments |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| ENP | Entry into the national phase |
Ref document number: 2018549816 Country of ref document: JP Kind code of ref document: A |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17842564 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 17842564 Country of ref document: EP Kind code of ref document: A1 |