ID 原文 译文
16915 最后通过改进的联合正交匹配追踪算法重构出目标3维图像。 Finally, the 3D imaging result isreconstructed by a improved joint Orthogonal Matching Pursuit (OMP) algorithm.
16916 实验结果表明，该方法具有较好的抗噪性能和成像质量，可以更好地反映目标外形几何特征。 The experimental resultsshow that the proposed method has good anti-noise and imaging quality, and can reflect the geometric detailsof the target.
16917 虹膜识别面临两个重要的问题：一是如何精细分解与重构虹膜球面图像； Iris recognition faces two important issues.
16918 二是如何识别虹膜图特征。 They are how to decompose finely and reconstruct the spherical image of the iris, and how to identify the characteristics of the iris.
16919 虹膜表面几何位置信息是一种重要的信号，传统的虹膜识别通常使用虹膜图像的平面特征，然而人的眼睛是一种球体，从平面图像难以提取到虹膜球体的几何特征。 Conventional iris recognition uses usually the planar features of these iris images. However, the human eye is a sphere. The geometric positioninformation of the iris surface is an important signal, but it is difficult to extract the geometric features of theiris sphere from the planar image.
16920 针对平面特征容易出现虹膜纹理的扭曲和失真等问题，该文建议一种正交对称的球面Haar小波(OSSHW)基，对球面虹膜信号进行多尺度分解与重构，获得更精细的虹膜曲面几何特征，同时对比球谐函数和半正交或正交球面Haar小波基的虹膜球面信号特征提取能力。 Considering the issue that the plane features are prone to distortion and lackfidelity of iris texture, an Orthogonal and Symmetric Spherical Haar Wavelet (OSSHW) basis is proposed todecompose and reconstruct the spherical iris signal to obtain stronger geometric features of iris surface. Thecomparison of the feature extraction ability to spherical signal by the spherical harmonics and the typical semi-orthogonal or nearly orthogonal spherical Haar wavelet is also presented.
16921 在此基础上，该文提出一种基于卷积神经网络(CNN)和正交对称的球面Haar小波的虹膜识别方法，它能够有效捕获虹膜球体曲面的局部精细特征，比半正交或正交球面Haar小波基具有更强的虹膜识别能力。 And then, an iris recognition methodbased on Convolutional Neural Networks (CNN) + OSSHW is proposed, which can effectively capture the localfine features of iris spherical surface, and has stronger ability in iris recognition than semi-orthogonal or nearlyorthogonal spherical Haar wavelet bases.
16922 Bisecting K-means算法通过使用一组初始中心对分割簇，得到多个二分聚类结果，然后从中选优以减轻局部最优收敛问题对算法性能的不良影响。 The algorithm of Bisecting K-means obtains multiple clustering results by using a set of initial centerpairs to segment a cluster, and then selects the best from them to mitigate the adverse effect of the localoptimal convergence on the performance of the algorithm.
16923 然而，现有的随机采样初始中心对生成方法存在效率低、稳定性差、缺失值等不同问题，难以胜任大数据聚类场景。 However, the current methods of random sampling togenerate initial center pairs for Bisecting K-means have some problems, such as low efficiency, poor stability,missing values and so on, which are not competent for big data clustering.
16924 针对这些问题，该文首先创建出了初始中心对组合三角阵和初始中心对编号三角阵， In order to solve these problems,firstly the lower triangular matrix composed by the pairs of initial centers and the lower triangular matrixcomposed by serial numbers of the pairs of initial centers are created.