共查询到18条相似文献,搜索用时 203 毫秒
1.
针对如何提高复杂曲面的三维模型的检索精度的问题,提出了一种基于曲度特征的三维模型检索算法。首先,在模型表面选取随机采样点,计算点所在局部曲面的高斯曲率和平均曲率,通过高斯曲率和平均曲率求出随机点的曲度值,曲度值表明了曲面的凹凸属性。然后,以模型的质心为球心,以随机点与质心距离和曲度值为坐标轴建立坐标系,统计出一定距离范围内曲度值分布的概率,构建距离与曲度的分布矩阵,以此分布矩阵作为三维模型特征描述符。该特征描述符具有旋转不变性和平移不变性,能够很好地反映复杂曲面的几何特征。最后,通过比较分布矩阵给出不同模型间的相似度。实验结果表明,该方法相比形状分布算法的检索性能有较大提高,尤其适用于具有复杂曲面的三维模型检索。 相似文献
2.
提出一种基于形状分布的机械CAD模型相似性评价算法.该方法首先在CAD模型表面随机取点,然后计算随机点之间的距离,通过对距离值进行统计得到形状分布曲线,最后通过对模型间形状分布曲线的比较得到两个模型的相似度,从而实现模型的相似性评价.实验表明该算法能较好的实现机械CAD模型的相似性评价. 相似文献
3.
4.
残缺的三维模型和完整的三维模型在拓扑结构上存在差异,为提高检索方法的鲁棒性,提出基于孔洞填充的残缺三维模型检索方法。通过基于径向基函数的孔洞填充方法对残缺三维模型进行孔洞填充,构造残缺模型对应的完整填充模型;构建填充模型中每个数据点的近邻矩阵,通过求解矩阵获得该点的曲率值;以聚类中心对每个数据点的影响为聚类标准改进[K]-menas算法,对填充模型的数据点进行聚类,引入C_DIS相似度量模型,对不同模型按类别进行匹配检索。实验结果表明,能有效检索出与待检索模型相似的三维模型。 相似文献
5.
6.
文中应用形状分布算法度量由CT数据重建出的三维骨骼的相似度,并提出了改进.基本步骤是:在模型表面选择随机点,计算每两点间的距离,以形状分布函数来构建形状分布直方图,通过比较形状分布直方图来给出不同骨骼间的相似度.实验表明,该方法能够准确度量骨骼的相似性. 相似文献
7.
8.
文中应用形状分布算法度量由CT数据重建出的三维骨骼的相似度,并提出了改进。基本步骤是:在模型表面选择随机点,计算每两点间的距离,以形状分布函数来构建形状分布直方图,通过比较形状分布直方图来给出不同骨骼间的相似度。实验表明,该方法能够准确度量骨骼的相似性。 相似文献
9.
10.
针对离散曲率估计对噪声敏感且特征值计算量大的特点提出了基于区域离散曲率的三维网格分水岭分割算法。寻找三维模型显著特征点;对三维模型进行预分割,确定分割带;在分割带区域上计算离散曲度极值点,利用测地距离和曲度极值点对三维模型进行分水岭分割。算法在分割前无需进行网格去噪,实验结果证明,对主体分支明显的模型具有较高的分割边缘准确度和较快的分割速度。 相似文献
11.
Freeform surfaces are popularly used to design and model complex 3D objects. These 3D models are stored as computerized models in databases. To facilitate data retrieval and shape matching, a major challenge lies in defining and computing the level of similarity between two or more freeform surfaces. In order to explore the useful 3D information associated with the surfaces, an integrated approach based on the integral of Gaussian curvature is proposed to develop the measures of similarity of freeform surfaces. Specifically, the integral of Gaussian curvature is mapped into the 2D space, and a shape-based measure is developed using statistical methods to compute the level of similarity. For smooth surfaces, a fast approximation algorithm is developed to calculate the curvature of individual subregions. In cases where the target surface has a complex topological structure or a smooth surface is not available, the integral of Gaussian curvature for the discrete surface is first calculated at each vertex, followed by mapping onto a 2D spherical coordinate. The distance measure focuses on the local geometry, which is critical to investigate models with a certain level of resemblance such as products in a family. This proposed approach can be applied to surfaces under various transformations, as well as 3D data from various sources. 相似文献
12.
3D image retrieval approach is a challenging problem in the research of content-based image retrieval. In this paper, a novel retrieval approach combined differential geometry and co-occurrence matrix is presented. Firstly, Gaussian curvature and mean curvature are utilized to represent the inherent characteristic of spatial surface, and then we use co-occurrence matrix to store the shape information of 3D images. Secondly, normalization process is applied to the co-occurrence matrix and the invariants independence of the translation, scaling, and rotation transforms are proved. In comparison with the recent methods, experiments indicate a lower computation complexity and a better retrieval rate to 3D images with slight different shape characteristic. 相似文献
13.
14.
15.
针对三维人脸数据庞大及识别效率低的问题,提出采用提取脊点及谷点表征人脸。脊点和谷点作为曲面局部区域内主曲率沿主方向变化的极值点,能够很好地表征三维人脸特征。对三维人脸提取脊点模型和谷点模型,通过对它们栅格化后生成对应的空间分布密度直方图实现人脸粗匹配,采用计算LTS-Hausdorff距离实现人脸的精确匹配。在GavabDB三维人脸库的实验结果表明,该方法具有较高的识别率。 相似文献
16.
Sketch-based 3D model retrieval is very important for applications such as 3D modeling and recognition. In this paper, a sketch-based retrieval algorithm is proposed based on a 3D model feature named View Context and 2D relative shape context matching. To enhance the accuracy of 2D sketch-3D model correspondence as well as the retrieval performance, we propose to align a 3D model with a query 2D sketch before measuring their distance. First, we efficiently select some candidate views from a set of densely sampled views of the 3D model to align the sketch and the model based on their View Context similarities. Then, we compute the more accurate relative shape context distance between the sketch and every candidate view, and regard the minimum one as the sketch-model distance. To speed up retrieval, we precompute the View Context and relative shape context features of the sample views of all the 3D models in the database. Comparative and evaluative experiments based on hand-drawn and standard line drawing sketches demonstrate the effectiveness and robustness of our approach and it significantly outperforms several latest sketch-based retrieval algorithms. 相似文献
17.
基于图像的三维模型检索比形状特征和拓扑结构特征具有更易于实现且检索效果好等优点,但现有基于图像的三维模型检索方法存在一些问题,如没有考虑三维深度信息、所提特征不能全部表达三维模型信息且不能实现用户交互操作等。对此提出一种基于深度图像的三维模型相关反馈检索算法:首先提取三维模型深度图像边界方向直方图和Zernike矩特征;然后利用特征距离度量进行三维模型检索,并采用相关反馈技术实现分类器的构造和模型的标注;最后利用调整好的权值分类器进行检索和标注。实验表明,该算法实现了三维模型的相关反馈检索和标注,提高了检索的精确性,并增强了系统的实用性,为进一步进行语义检索打下基础。 相似文献
18.
如何对三维模型进行特征提取是近年来出现的三维模型检索中的主要问题.文章给出了一种基于视点距离的特征提取算法,该算法利用正规化后的三维模型表面到观察点的距离信息生成六幅距离图像,然后对图像进行二维傅立叶变换并对变换后的频域信息进行低频采样从而得到三维模型的特征向量.该算法克服了基于三维投影的二维图像轮廓算法中丢失模型空域信息、缺乏对图像内部信息进行描述的缺点.实验结果表明,该算法比基于轮廓算法的检索精确度提高了19%. 相似文献