基于形变估计与运动补偿的医学CT图像层间超分辨率算法

针对医学断层图像层间分辨率较低的问题,提出了基于形变估计与运动补偿的医学CT图像层间超分辨率算法用于生成切片间图像,从而提高层间分辨率。首先利用U-Net对相邻两幅图像作多尺度特征提取与融合;其次,为了处理层间图像的复杂形变,使用基于自适应协作流的变形扭曲模块来实现相邻切片间的双向形变估计,设计层级信息递进融合模块对金字塔特征层进行特征聚合,对生成图进行运动补偿;最后经过后处理网络以减少异常像素点。该算法在两种CT数据集上进行验证,平均PSNR值分别达到了35.59 dB和30.76 dB,输出图能较好地恢复图像细节。与现有的一些方法对比,相关实验证明了该算法的有效性。

基于特征缺失修复的人脸三维运动形变估计算法

人脸由于其非刚性明显,受到运动背景变化、形变程度复杂、特征丢失等因素的影响,使得在进行三维动态人脸形变估计时,面临着估计不准,误差较大的问题。为了解决这些问题,提出一种基于特征缺失修复的人脸三维运动形变估计算法,将非刚体人脸运动的形变估计过程中,运用特征补偿方法,补偿由于噪声、形变程度复杂带来的特征丢失缺陷,运用较多的人脸形变特征进行形变分析,准确计算人脸的形变程度。实验结果表明,这种方法得到的解误差较小,效果明显。

附加系统参数的多时相InSAR时空建模和形变估计

轨道误差和长波大气延迟组成的系统误差是影响In SAR形变监测精度的重要因素之一.传统方法在空间域对干涉图的系统误差建模,容易导致长波形变和系统误差相混淆.本文在时空域利用附加系统参数对系统误差建模,同时根据观测值质量对差分相位观测值定权,采用附加系统参数的加权最小二乘法估计形变参数和系统误差,实现了长波形变和系统误差的分离.模拟实验结果表明,在形变与系统误差的空间变化特性完全一致的极端情况下,本文方法能实现两者的有效分离,估计的形变速率均方根误差比传统方法降低了98. 8%.ASAR数据实验显示当形变尺度较小且分散分布时,本文方法和传统方法得到的结果相似;当形变在研究区内表现为长波变化时,本文方法比传统方法估计的形变结果更为稳健.

一种基于随机图像序列的三维人脸识别算法

针对二维图像序列非刚体三维特征恢复识别技术受到背景多样性、运动复杂性、数据的丢失、特征点噪声等问题的影响,识别重建人脸特征面临着形变程度估计不准,识别误差较大的问题,提出一种基于随机图像序列的三维人脸识别算法。该方法通过在三维人脸运动恢复的框架模型中加入一定的物理运动约束,刻画不同非刚体运动形状基的情况下,准确计算人脸运动形变的程度,运用形状基模糊关联技术完成人脸的三维识别。实验结果表明,这种方法得到的解误差较小,识别效果明显。

多模态医学图像配准算法综述

以多模态医学影像(多序列磁共振(MR)、计算机断层扫描(CT)和X光等模态)为研究对象,对近10 a的多模态医学影像配准相关研究工作进行归纳和分析.首先阐述面向临床应用的多模态医学配准的必要性,分析影像配准的一般流程;然后提出三种归纳角度对现有研究进行分析,重点总结多序列MR、CT和X光等模态之间的五种配准模式、五类常见配准的解剖结构及四类常用的多模态影像配准算法;接着分析七个用于多模态医学影像配准数据集和六个常用的配准评价指标;最后指出多模态医学影像配准算法面临的挑战和未来趋势.

Estimation of deformed laser heat sources and thermal analysis on laser assisted turning of square member

Laser assisted machining （LAM） has difficulties in estimating temperature after applying a LAM process due to its very small heat input area, large energy and movement. In particular, in the case of laser assisted turning （LAT） process, it is more difficult to estimate the temperature after preheating because it has a shape of ellipse when a laser heat source is rotated. A prediction method and thermal analysis method for heat source shapes were proposed as a square shaped member was preheated. The temperature distribution was calculated according to the rotation of the member. Compared with the results of the former study, the maximum temperature of the calculation results, 1 407.1 ℃, is 8.5 ℃ higher than that of the square member, which is 1 398.6 ℃. In a LAT process for a square member, the maximum temperature is 1 850.8 ℃. It is recognized that a laser power control process is required because square members show a maximum temperature that exceeds a melting temperature at around a vertex of the member according to the rotation.

Volume Estimation of Small Scale Debris Flows Based on Observations of Topographic Changes Using Airborne LiDAR DEMs

This paper describes a geographic information system(GIS)-based method for observing changes in topography caused by the initiation, transport, and deposition of debris flows using highresolution light detection and ranging(LiDAR) digital elevation models(DEMs) obtained before and after the debris flow events. The paper also describes a method for estimating the volume of debris flows using the differences between the LiDAR DEMs. The relative and absolute positioning accuracies of the LiDAR DEMs were evaluated using a real-time precise global navigation satellite system(GNSS) positioning method. In addition, longitudinal and cross-sectional profiles of the study area were constructed to determine the topographic changes caused by the debris flows. The volume of the debris flows was estimated based on the difference between the LiDAR DEMs. The accuracies of the relative and absolute positioning of the two LiDAR DEMs were determined to be ±10 cm and ±11 cm RMSE, respectively, which demonstrates the efficiency of the method for determining topographic changes at an scale equivalent to that of field investigations. Based on the topographic changes, the volume of the debris flows in the study area was estimated to be 3747 m3, which is comparable with the volume estimated based on the data from field investigations.

MODELING INTRAPERSONAL DEFORMATION SUBSPACE USING GMM FOR PALMPRINT IDENTIFICATION

In this paper, an efficient model of palmprint identification is presented based on subspace density estimation using Gaussian Mixture Model (GMM). While a few training samples are available for each person, we use intrapersonal palmprint deformations to train the global GMM instead of modeling GMMs for every class. To reduce the dimension of such variations while preserving density function of sample space, Principle Component Analysis (PCA) is used to find the principle differences and form the Intrapersonal Deformation Subspace (IDS). After training GMM using Expectation Maximization (EM) algorithm in IDS, a maximum likelihood strategy is carried out to identify a person. Experimental results demonstrate the advantage of our method compared with traditional PCA method and single Gaussian strategy.

A CODEBOOK COMPENSATIVE VOICE MORPHING ALGORITHM BASED ON MAXIMUM LIKELIHOOD ESTIMATION

This paper presents an improved voice morphing algorithm based on Gaussian Mixture Model(GMM) which overcomes the traditional one in the terms of overly smoothed problems of the converted spectral and discontinuities between frames.Firstly, a maximum likelihood estimation for the model is introduced for the alleviation of the inversion of high dimension matrixes caused by traditional conversion function.Then, in order to resolve the two problems associated with the baseline, a codebook compensation technique and a time domain medial filter are applied.The results of listening evaluations show that the quality of the speech converted by the proposed method is significantly better than that by the traditional GMM method, and the Mean Opinion Score(MOS) of the converted speech is improved from 2.5 to 3.1 and ABX score from 38% to 75%.

时序InSAR对流层大气延迟改正的相位堆叠方法

对流层大气延迟一直是限制合成孔径雷达干涉测量InSAR(Synthetic Aperture Radar Interferometry)技术形变测量精度的主要因素之一。基于含有共同日期的干涉相位也包含相同的大气延迟相位分量这一事实,相位堆叠CSS(Common Scene Stacking)方法采用叠加求和的方式来估计共有日期的大气延迟相位,并通过迭代运算提高估计精度,具有实现简单、计算效率高的优势。本文对CSS方法估算出的大气延迟相位进行空间低通滤波,并将相位解缠与CSS大气延迟相位估计进行迭代运算,改善形变估计结果。首先,基于模拟数据分析了迭代次数和时间窗口尺寸对CSS方法对流层大气延迟改正效果的影响;然后,将本文方法应用于真实SAR数据,分析了不同大小的时间窗口对改正结果的影响,并与其他方法进行对比,结果表明该方法可以获得较好的大气延迟改正结果,改正后空间相位标准差平均降低了62%,稳定点的时间序列标准差平均降低了58%。在此基础上,对目前存在的问题进行了分析和讨论。

Direct estimation of rupture depths of earthquake faults from coseismic surface deformation

The rupture dimensions of earthquake faults are important parameters for characterizing earthquake ruptures and ground motions. Two key parameters to be determined are the rupture depth and dip angle of earthquake faults. Dislocation theory in an elastic half space indicates that if a seismic rupture directly runs up to the ground surface, there exist zero points of horizontal strain in the surface deformation, which correspond to the rupture depths, except for pure strike-slip faults. In this study, we use numerical simulations to investigate the possibility of inferring rupture depths from zero-strain points for cases of buried faults and heterogeneous media. The results show that the correspondence of zero-strain points to the rupture depths can be influenced by the heterogeneity of the underground media and the stress field. For buried faults, the correspondence relationship is approximately valid when the fault depth is <1 km. In addition, the range of earthquake fault dip angles can be estimated by horizontal displacements on the ground. We also study how to determine the rupture depths of faults from InSAR data after large earthquakes, and successfully apply the method to the 2008 Wenchuan earthquake. The method proposed here, which determines the parameters of fault geometry according to surface deformation, is simple and easy to perform. With independent of aftershocks, it can provide valuable constraints to kinematic inversions.

GEOMETRY OF COMPLETE HYPERSURFACES EVOLVED BY MEAN CURVATURE FLOW

Some geometric behaviours of complete solutions to mean curvature flow before the singu-larities occur are studied. The author obtains the estimates of the rate of the distance betweentwo fixed points and the derivatives of the second fundamental form. By use of a new maximumprinciple, some geometric properties at infinity are obtained.

THEOREMS OF BARTH-LEFSCHETZ TYPE ON KAHLER MANIFOLDS WITH PARTIALLY POSITIVE CURVATURE

The author obtains the theorems of Barth-Lefschetz type on Kahler manifolds with partiallypositive bisectional curvature without the assumption of nonnegative bisectional curvature.Some applications of the results to holomorphic mappings are given.

Identification of flexible buildings with bending deformation and the unmeasured earthquake ground motion

It is important to explore efficient algorithms for the identification of both structural parameters and unmeasured earthquake ground motion.Recently,the authors proposed an algorithm for the identification of shear-type buildings and unknown earthquake excitation.In this paper,it is extended to the investigation of the identification of flexible buildings with bending deformation and the unmeasured earthquake ground motion.In the absolute co-ordinate system,the unmeasured ground motion can be treated as an unknown translational force and a bending moment at the 1st floor level of a flexible building.Structural unknown parameters above the 1st story of the building can be identified by the extended Kalman estimator and the 1st story stiffness and the unmeasured ground motion are subsequently estimated based on the least-squares estimation.The proposed algorithm is further extended to the identification of tall bending-type buildings based on substructure approach.Inter-connection effect between sub-buildings is treated as‘additional unknown inputs’to sub-buildings,which are estimated by the extended Kalman estimator without the measurements of rotational responses.Numerical examples demonstrate the identification of a multi-story,tall bending-type building and its unmeasured earthquake ground motions using only partial measurements of structural absolute responses.