从运动中恢复目标结构的改进因子分解法

因子分解法是从图像序列中恢复刚体目标几何结构的重要方法。针对传统因子分解法基本过程中存在的不足,及其容易失效的缺点,提出一种改进的因子分解法。该方法避开传统方法中求解修正矩阵的复杂过程,利用旋转矩阵的特性,直接修正由传统方法奇异值分解(SVD)得到的每帧图像的旋转矩阵,然后根据观测矩阵和得到的旋转矩阵,直接利用线性最小二乘法求解目标的结构矩阵。仿真和实测数据的实验结果表明,本文方法能够有效地从序列图像中恢复目标的几何结构,相比传统因子分解法,在稳定性上有较大的提升。     

基于迭代张量高阶奇异值分解的运动目标提取

将基于凸优化的低秩矩阵恢复(LRMR)理论用于背景建模,当背景不稳定时,这种方法提取运动目标的效果不佳。由于矩阵的数据表示形式破坏了视频在时间和空间上的原始结构,本文采用张量表征视频的高维结构特性,提出了一种基于迭代张量高阶奇异值分解(HOSVD)的运动目标提取方法。本文用高阶奇异值分解代替LRMR中的矩阵奇异值分解(SVD),利用增广拉格朗日乘子法重建出三维视频张量的背景部分和运动目标部分,并进一步对运动目标部分进行形态学开闭运算。实验结果证明,相比常用方法, 本文方法错分率更低,能更准确完整地提取运动目标。     

基于运动恢复结构方法的植株三维重建

为了探究基于运动恢复结构(Structure From Motion, SFM)方法的植株三维重建模型的效果,为植物三维重建工作提供研究案例,本文以紫叶鸭跖草(setcreasea pallida)为研究对象,在搭建序列图像获取平台的基础上,选取35幅、75幅、105幅序列图像进行三维重建的对比分析;同时从植株表型参数方面,对植株三维重建模型进行精度评价。结果表明：75幅图像序列的重建效果最好;不同图像序列的模型计算的植株高度相对误差（Relative Error, RE）均小于2.5%,决定系数（coefficient of determination, R2）均大于0.998;不同图像序列的模型提取叶片长和叶片宽的RE均小于2.89%,R2均大于0.958。因此,序列图像的数量与重建模型的效果有关,但二者并非呈正相关关系;序列图像的数量对重建叶片的长与宽的误差影响较小;SFM方法应用于结构比较复杂的植株的三维重建可以取得较好的重建效果。     

非凸加权核范数及其在运动目标检测中的应用

目的 近年来,低秩矩阵分解被越来越多的应用到运动目标检测中。但该类方法一般将矩阵秩函数松弛为矩阵核函数优化,导致背景恢复精度不高;并且没有考虑到前景目标的先验知识,即区域连续性。为此提出一种结合非凸加权核范数和前景目标区域连续性的目标检测算法。方法 本文提出的运动目标检测模型以鲁棒主成分分析(RPCA)作为基础,在该基础上采用矩阵非凸核范数取代传统的核范数逼近矩阵低秩约束,并结合了前景目标区域连续性的先验知识。该方法恢复出的低秩矩阵即为背景图像矩阵,而稀疏大噪声矩阵则是前景目标位置矩阵。结果 无论是在仿真数据集还是在真实数据集上,本文方法都能够取得比其他低秩类方法更好的效果。在不同数据集上,该方法相对于RPCA方法,前景目标检测性能提升25%左右,背景恢复误差降低0.5左右;而相对于DECOLOR方法,前景目标检测性能提升约2%左右,背景恢复误差降低0.2左右。结论 矩阵秩函数的非凸松弛能够比凸松弛更准确的表征出低秩特征,从而在运动目标检测应用中更准确的恢复出背景。前景目标的区域连续性先验知识能够有效地过滤掉非目标大噪声产生的影响,使得较运动目标检测的精度得到大幅提高。因此,本文方法在动态纹理背景、光照渐变等较复杂场景中均能够较精确地检测出运动目标区域。但由于区域连续性的要求,本文方法对于小区域多目标的检测效果不甚理想。     

从二次曲线对应恢复刚体三维运动和结构

研究从二次曲线恢复刚体的三维运动参数和结构参数问题.建立了基于二次曲线的几何模型和运动模型,从理论上证明最少需要3条对应的二次曲线,建立一个由6个非线性方程构成的非线性方程组.能求出刚体的旋转运动参数和包含一个比例因子(不等于零)的平移运动参数.在此基础上得到空间二次曲线的参数,对应于每一组运动参数,空间二次曲线的参数有两个解.     

图像序列的增量式运动结构恢复

目的 传统增量式运动结构恢复算法中,初始图像对选择鲁棒性差,增量求解过程效率较低,捆绑调整策略存在计算冗余,模型修正后仍存在较大误差。为解决上述问题,以基于图像序列的3维重建为基础,提出一种新的增量式运动结构恢复算法（SFM-Y）。方法 首先,采用改进的自适应异常值过滤方法增强初始图像对选择的鲁棒性,得到用于初始重建的初始图像对;其次,通过增量迭代重建丰富点云模型,采用改进的EPNP（efficient perspective-n-point）解算方法提高增量添加过程的计算效率和精确度;最后,采用优化的捆绑调整策略进行模型修正,解决模型漂移问题,修正重投影误差。结果 实验选取不同数据规模的数据集,在本文方法及传统方法间进行测试对比,以便更加全面地分析算法性能。实验结果表明,SFM-Y算法相比传统的增量式运动结构恢复算法,在计算效率和结果质量方面均有所提高,根据性能分析对比的结果所示,本文方法较传统方法在计算效率和重建精度上约有10%的提升。结论 提出的增量式运动结构恢复算法能够高效准确地实现基于图像序列的3维重建优于传统方法,计算效率较高,初始重建鲁棒性强,生成模型质量较好。     

基于低秩矩阵分解的运动目标检测

运动目标检测是视频监控任务的基础问题之一,针对灰度信息,目标检测存在的阴影识别能力差、检测精度低等问题,提出在HSV颜色空间下基于低秩矩阵分解的运动目标检测算法.首先将获取的RGB图像转为HSV颜色空间分量,分别对H、S、V通道构建低秩观测量,进行低秩矩阵优化分解,分离出不同颜色通道的前景和背景分量;组合H、S、V通道分量的前景图像,得到粗略的运动目标区域;再采用HSV颜色阴影去除去除前景图像中的阴影;最后经噪声去除和空洞的填充,检测得到准确的前景运动目标.实验验证表明,与其它方法相比,能够有效地提高运动目标检测的准确度.     

基于分段式序列图片集的运动恢复结构

多视图运动恢复结构（Structure from Motion,SFM）是三维重建中相机姿态估计的一种最常用的方法。传统SFM采用增量方式处理图片,算法的时间复杂度是[O(n4)],当图片数量较多时,重建时间很长。此外,由于图片噪声影响,漂移误差将随着图片数量增加不断累加,影响最终的重建质量。添加集束调整（Bundle Adjustment,BA）可以优化重建结果,但是需要花费更长的时间。在现有增量式算法的基础上,提出基于分段式序列图片集的方法,将序列图片集按照相似度划分为小集合,对每个小集合进行并行计算,减少误差累积量和重建时间,最后再用BA进行全局优化。实验结果表明,该方法能在保持一定精度的前提下,有效减少重建时间。     

非零边界旋转运动模糊图像的恢复算法

具有非零灰度边界的旋转运动模糊图像的恢复是图像处理的一个热点研究问题。为了在离散情况下能够沿模糊路径上快速提取像素的灰度信息，将Bresenham算法思想引入到旋转运动模糊图像的恢复工作中，并在此基础上，提出了一种基于模糊路径频域滤波去卷积的恢复算法，同时，为了有效地恢复非零边界区的图像信息，又提出了一种基于邻域知识引导的最小代价恢复算法。这样就可将非零边界旋转运动模糊图像的恢复过程分为两个阶段，即先后用该两个恢复算法分别进行处理，由于使整个恢复过程避开了迭代计算和一些复杂的费时运算，从而加快了恢复速度。为验证该算法的效果，在微机上，对该恢复算法与现有的算法进行了对比实验和一系列的验证实验，实验结果表明，该算法十分有效，且具有一定的抗噪能力。     

基于图像序列的非刚体三维运动恢复

研究图像序列中非刚体的三维运动重建问题。介绍非刚体运动重建的两种重要算法:奇异值分解法和线性迭代法。对迭代算法所用的重构方法进行修改,在迭代过程中应用结果更为精确的重构方法来求解非刚体的模型和旋转矩阵。对真实图像序列的实验结果验证了该算法的有效性和精确性。     

An Iterative Multiresolution Scheme for SFM with Missing Data

Several techniques have been proposed for tackling the Structure from Motion problem through factorization in the case of missing data. However, when the percentage of unknown data is high, most of them may not perform as well as expected. Focussing on this problem, an iterative multiresolution scheme, which aims at recovering missing entries in the originally given input matrix, is proposed. Information recovered following a coarse-to-fine strategy is used for filling in the missing entries. The objective is to recover, as much as possible, missing data in the given matrix. Thus, when a factorization technique is applied to the partially or totally filled in matrix, instead of to the originally given input one, better results will be obtained. An evaluation study about the robustness to missing and noisy data is reported. Experimental results obtained with synthetic and real video sequences are presented to show the viability of the proposed approach.

A semi-direct approach to structure from motion

The problem of structure from motion is often decomposed into two steps: feature correspondence and three-dimensional reconstruction. This separation often causes gross errors when establishing correspondence fails. Therefore, we advocate the necessity to integrate visual information not only in time (i.e. across different views), but also in space, by matching regions – rather than points – using explicit photometric deformation models. We present an algorithm that integrates image-feature tracking and three-dimensional motion estimation into a closed loop, while detecting and rejecting outlier regions that do not fit the model. Due to occlusions and the causal nature of our algorithm, a drift in the estimates accumulates over time. We describe a method to perform global registration of local estimates of motion and structure by matching the appearance of feature regions stored over long time periods. We use image intensities to construct a score function that takes into account changes in brightness and contrast. Our algorithm is recursive and suitable for real-time implementation.     

A neural model for the integration of stereopsis and motion parallax in structure from motion

We introduce a model for the computation of structure from motion based on the physiology of visual cortical areas MT and MST. The model assumes that the perception of depth from motion is related to the firing of a subset of MT neurons tuned to both velocity and disparity. The model's MT neurons are connected to each other laterally to form modulatory receptive-field surrounds that are gated by feedback connections from area MST. This allows the building up of a depth map from motion in area MT, even in absence of disparity in the input. Depth maps from motion and from stereo are combined by a weighted average at a final stage. The model's predictions for the interaction between motion and stereo cues agree with previous psychophysical data, both when the cues are consistent with each other or when they are contradictory. In particular, the model shows nonlinearities as a result of early interactions between motion and stereo before their depth maps are averaged. The two cues interact in a way that represents an alternative to the “modified weak fusion” model of depth–cue combination.     

An iterative multiresolution scheme for SFM with missing data: Single and multiple object scenes

Most of the techniques proposed for tackling the Structure from Motion problem (SFM) cannot deal with high percentages of missing data in the matrix of trajectories. Furthermore, an additional problem should be faced up when working with multiple object scenes: the rank of the matrix of trajectories should be estimated. This paper presents an iterative multiresolution scheme for SFM with missing data to be used in both the single and multiple object cases. The proposed scheme aims at recovering missing entries in the original input matrix. The objective is to improve the results by applying a factorization technique to the partially or totally filled in matrix instead of to the original input one. Experimental results obtained with synthetic and real data sequences, containing single and multiple objects, are presented to show the viability of the proposed approach.     

Preemptive RANSAC for live structure and motion estimation

A system capable of performing robust live ego-motion estimation for perspective cameras is presented. The system is powered by random sample consensus with preemptive scoring of the motion hypotheses. A general statement of the problem of efficient preemptive scoring is given. Then a theoretical investigation of preemptive scoring under a simple inlier–outlier model is performed. A practical preemption scheme is proposed and it is shown that the preemption is powerful enough to enable robust live structure and motion estimation.Prepared through collaborative participation in the Robotics Consortium sponsored by the U.S. Army Research Laboratory under the Collaborative Technology Alliance Program, Cooperative Agreement DAAD19-01-2-0012. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation thereon. David Nistér received PhD degree in computer vision, numerical analysis and computing science from the Royal Institute of Technology (KTH), Stockholm, Sweden, with the thesis ‘Automatic Dense Reconstruction from Uncalibrated Video Sequences’. He is currently an assistant professor at the Computer Science Department and the Center for Visualization and Virtual Environments, University of Kentucky, Lexington. Before joining UK, he was a researcher in the Vision Technologies Laboratory, Sarnoff Corporation, Princeton, and Visual Technology, Ericsson Research, Stockholm, Sweden. His research interests include computer vision, computer graphics, structure from motion, multiple view geometry, Bayesian formulations, tracking, recognition, image and video compression. He is a member of the IEEE and American Mensa.     

Discontinuity pattern detection and orientation measurement for tunnel faces by using structure from motion photogrammetry

Crack is a common condition that affects water-rich loess tunnels, and it negatively influences the reliability and safety of the tunnel. Jointed rock masses are commonly found during underground excavation. Many underground openings have failed during excavation and operation. The evaluation of tunnel faces by using a new measurement system based on structure from motion (SfM) photogrammetry is proposed in this study. The main objective is to determine the discontinuity pattern and orientation of tunnel faces as an input for tunnel face stability evaluation. A set of overlapping images obtained from a tunnel face replica using 3D polystyrene and the SfM photogrammetry approach is utilized to generate a 3D point cloud model. The discontinuity pattern and orientation are determined via facet extraction of the K_D-tree plugin in CloudCompare. The same set of overlapping images is analyzed in four different quality settings (low, medium, high, and ultrahigh) with different sets of point cloud numbers that control the accuracy of the measurement of discontinuity. Results show that the high-quality setting presents a consistent measurement of the discontinuity pattern and orientation in contrast to the real 3D polystyrene tunnel face model. 2D pattern results from CloudCompare are validated using the fractal analysis method in an image analysis software, where the image is converted into a binary-one. Two sets of discontinuity are derived. The means of dip and dip direction from the cluster analysis are 82°/164° and 84°/307°, respectively. The orientations are verified through manual compass measurement. The orientations for manual compass measurement for Sets 1 and 2 are 86°/151° and 80°/303°, respectively. The discontinuities obtained from the high-quality setting of the point cloud and manual measurement of the orientation by using a geological compass highlight the similarity of the discontinuity plane in both discontinuity sets for the tunnel face replica. The proposed method for tunnel face evaluation has good judgment for tunnel support and the prevention of biased analyses by professionals.     

An improved motion compensation method for high resolution UAV SAR imaging

The polynomial and sinusoidal motion errors always exist in the unmanned aero vehicle （UAV） SAR due to the small size and low velocity of the platform, causing serious spectrum compressing/stretching and significant spectral replicas of the azimuth signal. The motion errors induce serious blurring of the SAR image and ＂ghost targets＂, and can hardly be precisely estimated by the conventional motion compensation （MOCO） method. In this paper, an improved MOCO method is proposed to estimate and eliminate the motion errors in the high resolution UAV SAR without high precision inertial navigation system （INS） data. The time domain range walk correction （RWC） operation in the coarse phase error estimation process of the proposed MOCO method is the key operation that ensures the estimation accuracy of the whole MOCO method. Finally, the validity of the improved MOCO method is verified by computer simulations and real UAV SAR data processing.     

Inferring 3D structure from image motion: The constraint of Poinsot motion

Monocular observers perceive as three-dimensional (3D) many displays that depict three points rotating rigidly in space but rotating about an axis that is itself tumbling. No theory of structure from motion currently available can account for this ability. We propose a formal theory for this ability based on the constraint of Poinsot motion, i.e., rigid motion with constant angular momentum. In particular, we prove that three (or more) views of three (or more) points are sufficient to decide if the motion of the points conserves angular momentum and, if it does, to compute a unique 3D interpretation. Our proof relies on an upper semicontinuity theorem for finite morphisms of algebraic varieties. We discuss some psychophysical implications of the theory.This work was supported by National Science Foundation grants IRI-8700924 and DIR-9014278 and by Office of Naval Research contract N00014-88-K-0354.     

An improved SVD-based watermarking technique for copyright protection

The drawbacks of SVD-based image watermarking are false positive, robust and transparency. The former can be overcome by embedding the principal components of the watermark into the host image, the latter is dependent on how much the quantity (i.e., scaling factor) of the principal components is embedded. For the existing methods, the scaling factor is a fixed value; actually, it is image-dependent. Different watermarks need the different scaling factors, although they are embedded in the same host image. In this paper, two methods are proposed to improve the reliability and robustness. To improve the reliability, for the first method, the principal components of the watermark are embedded into the host image in discrete cosine transform (DCT); and for the second method, those are embedded into the host image in discrete wavelets transform (DWT). To improve the robustness, the particle swarm optimization (PSO) is used for finding the suitable scaling factors. The experimental results demonstrate that the performance of the proposed methods outperforms than those of the existing methods.