局部相位相关用于图像亚像素级配准技术研究

提出了一种基于局部相位相关的高效和鲁棒的亚像素级图像配准方法。通过传统的相位相关算法估计出初始平移参数后,在初始位置的引导下对互相关功率谱进行上采样矩阵Fourier变换,实现了图像局部相位相关,得到图像间亚像素级平移参数。实验结果表明,算法配准精度较高,且对随机噪声和光照变化具有较强的鲁棒性。     

Scale-Free Registrations in 3D: 7 Degrees of Freedom with Fourier Mellin SOFT Transforms

Fourier Mellin SOFT (FMS) as a novel method for global registration of 3D data is presented. It determines the seven degrees of freedom (7-DoF) transformation, i.e., the 6-DoF rigid motion parameters plus 1-DoF scale, between two scans, i.e., two noisy, only partially overlapping views on objects or scenes. It is based on a sequence of the 3D Fourier transform, the Mellin transform and the SO(3) Fourier transform. This combination represents a non-trivial complete 3D extension of the well known Fourier-Mellin registration for 2D images. It is accordingly based on decoupling rotation and scale from translation. First, rotation—which is the main challenge for the extension to 3D data - is tackled with a SO(3) Fourier Transform (SOFT) based on Spherical Harmonics. In a second step, scale is determined via a 3D Mellin transform. Finally, translation is calculated by Phase-Matching. Experiments are presented with simulated data sets for ground truth comparisons and with real world data including object recognition and localization in Magnetic Resonance Tomography (MRT) data, registration of 2.5D RGBD scans from a Microsoft Kinect with a scale-free 3D model generated by Multi-View Vision, and 3D mapping by registration of a sequence of consecutive scans from a low-cost actuated Laser Range Finder. The results show that the method is fast and that it can robustly handle partial overlap, interfering structures, and noise. It is also shown that the method is a very interesting option for 6-DoF registration, i.e., when scale is known.     

基于核相关滤波器的目标旋转跟踪算法

提出一种新的鲁棒的旋转估计算法,以核相关滤波器理论为基础,通过在目标中心等角度间隔来采样一个样本金字塔,单独训练一个一维的角度估计滤波器,从而将目标旋转角估计问题变为一个检测问题.提出的角度估计方法具有通用性,可以辅助其它没有角度估计功能的跟踪器.在实验中,单独训练一个位移跟踪滤波器和一个尺度估计滤波器,结合本文提出的角度估计滤波器形成一个由三层滤波器组成的跟踪器.和经典的算法在不同测试数据上的对比实验表明,该算法能达到较高的跟踪精度.     

结合Zernike矩和模板具有RST不变性的DWT-HMM鲁棒水印算法

抵抗几何攻击是鲁棒水印研究的关键问题之一。为实现水印同步,提出了一种结合Zernike矩和小波域模板实现由粗到精几何同步的算法,即首先利用平移归一化图像的Zernike矩估计旋转和缩放参数,并用校正旋转和缩放后的图像和原图像之间的质心增量估计平移参数;然后基于粗略估计的参数,通过匹配小波域模板实现旋转、缩放和平移(RST)参数的精确识别和校正,该同步方法可以较大程度地降低搜索空间。水印嵌入和检测采用了小波域向量隐马尔可夫模型(DWT-HMM)。仿真结果表明,利用由粗到精的几何同步方法和基于HMM的水印算法能有效抵抗StirMark平台的多种单项攻击和联合攻击,算法具有较好的鲁棒性能。     

A frequency domain technique for range data registration

This work introduces an original method for registering pairs of 3D views consisting of range data sets which operates in the frequency domain. The Fourier transform allows the decoupling of the estimate of the rotation parameters from the estimate of the translation parameters, our algorithm exploits this well-known property by suggesting a three-step procedure. The rotation parameters are estimated by the first two steps through convenient representations and projections of the Fourier transforms' magnitudes and the translational displacement is recovered by the third step by means of a standard phase correlation technique after compensating one of the two views for rotation. The performance of the algorithm, which is well-suited for unsupervised registration, is clearly assessed through extensive testing with several objects and shows that good and robust estimates of 3D rigid motion are achievable. Our algorithm can be used as a prealignment tool for more accurate space-domain registration techniques, like the ICP algorithm.     

基于Zernike矩和前馈神经网络的图像配准

提出一种基于Zernike矩和多级前馈神经网络的图像配准算法。利用低阶Zernike矩表征图像的全局几何特征,通过多级前馈神经网络学习图像所经历的旋转、缩放和平移等仿射变换参数,在一级前馈神经网络的基础上添加二级前馈网络,以提高参数估计精度。仿真结果表明,与基于DCT系数的神经网络算法相比,该算法旋转、缩放和平移估计精度较高,对噪声的鲁棒性较强。     

一种融合纹理的三维图像重建快速实现方法

为了得到完整的三维模型,介绍了一种融合纹理的三维图像重建快速实现方法。通过对不同视角的深度图像的手动粗配准、ICP算法精配准以及全局配准得到这些深度图像的旋转平移矩阵。通过vrippack,三维重建出完整的三维图像,用TextureStitcher对得到的三维图像进行纹理映射,从而实现融合纹理的三维图像的快速重建。文中在论述配准算法主要思想和实现步骤的同时,也用实验验证了方法的可行性与通用性。     

Recovery of ego-motion using region alignment

A method for computing the 3D camera motion (the ego-motion) in a static scene is described, where initially a detected 2D motion between two frames is used to align corresponding image regions. We prove that such a 2D registration removes all effects of camera rotation, even for those image regions that remain misaligned. The resulting residual parallax displacement field between the two region-aligned images is an epipolar field centered at the FOE (Focus-of-Expansion). The 3D camera translation is recovered from the epipolar field. The 3D camera rotation is recovered from the computed 3D translation and the detected 2D motion. The decomposition of image motion into a 2D parametric motion and residual epipolar parallax displacements avoids many of the inherent ambiguities and instabilities associated with decomposing the image motion into its rotational and translational components, and hence makes the computation of ego-motion or 3D structure estimation more robust     

基于边缘特征和Keren算法的图像配准

为获得高精确度、高稳定度的岩心配准图像,提出利用Sobel算子提取图像边缘并结合keren算法的图像配准方法。该方法首先提取参考图像和待配准图像边缘,接着对边缘图像进行高斯金字塔分解,最后,利用keren改进算法实现由粗到细的岩心图像配准。实验结果表明该算法与Keren算法相比,在小角度下都有较高的配准精度,大角度旋转情况下该方法绝对误差明显降低,且在大角度旋转下平移参数和角度参数分别可获得0.1个像素以下和0.1度以下的绝对误差精度,提高了岩心配准精度。     

Radon变换域的鲁棒图像Hash算法研究

针对数字图像内容认证的问题,提出了一种基于Radon变换域的鲁棒图像Hash算法。该算法在Radon变换域提取图像平移、缩放不变的矩特征,将矩特征的离散傅里叶变换系数作为图像的Hash值。离散傅里叶变换可实现抗图像旋转操作,并可降低图像Hash的维数。为了增加安全性,整个特征提取过程都是基于密钥进行的。理论分析证明Radon变换域的不变矩特征对一般的图像处理操作,特别是平移、缩放、旋转等几何操作有极好的不变性。实验结果表明提出的算法几乎可以容忍所有的图像内容保存操作,比如缩放、旋转、JPEG压缩、滤波和加噪等,同时对视觉上有明显失真或根本不相似的图像有高度的敏感性。     

Visible and infrared image registration based on region features and edginess

Visible and infrared image registration is required for multi-sensor fusion and cooperative processing. However, traditional single-sensor image registration methods are generally not feasible as multi-sensor images are often loosely related and show different properties in imaging. This paper presents a coarse-to-fine procedure for registering visible and infrared images based on stable region features and edginess. Zernike moments are used to describe salient region features for a coarse registration, and an entropy optimal process based on edginess is used to refine the registration to achieve a more accurate result. Experiments show that the proposed method provides more robust and accurate registration than the existing methods.     

一种无标记点三维点云自动拼接技术

基于相位投影和双目的三维光学测量系统已经广泛应用于各领域.受投影光测量系统单次测量范围大小的限制,对大型物体的测量需要在表面粘贴圆形标记点进行多次拼接的缺点,探讨一种基于SIFT的无标记点自动拼接技术.该技术采用SIFT方法获取两次测量的特征点,其次结合RANSAC求出图像特征点的匹配关系,再根据立体匹配中图像特征点与三维点云之间的对应关系,将二维特征点的对应关系映射到三维点云的对应关系上,最后由SVD奇异值分解算法求得旋转和平移矩阵实现拼接.实验证明:该方法可以避免在被测量对象上粘贴标记点,能够快速准确地实现自动拼接.     

Efficient registration of 2D points to CAD models for real-time applications

Efficient registration is a major challenge for real-time machine vision applications. Modern acquisition hardware can produce data at extremely high rates. Thus, efficient registration algorithms are required to align data to reference models to detect deviations and take correcting actions if needed. In this work, an efficient registration procedure of 2D points to CAD (computer-aided design) models is proposed. Recent developments in the field are reviewed and evaluated in terms of accuracy, speed and robustness. Efficient algorithms are proposed for the most computationally expensive parts of the registration, including an estimation of the rigid transform, a calculation of the closest point to geometric primitives, and an estimation of the surface normal. Furthermore, a novel primitive caching procedure is proposed that, when combined with an R-tree, greatly improves the execution speed of the registration. The result is a very accurate registration procedure, since geometric primitives are treated analytically with no point sampling required. At the same time, the proposed procedure is robust, very fast, and can achieve the correct registration in less than one millisecond.     

基于改进的SURF算子和透视变换模型的图像配准

对于传统的图像匹配算法存在特征信息少、错误匹配率高的问题,提出了一种基于改进的SURF算子和通过透视变换模型的图像配准算法。首先对传统的SURF描述符进行改进来进行特征点检测,然后用FLANN(Fast Library for Approximate Nearest Neighbors)搜索算法对检测出的角点进行粗匹配,再采用随机抽样一致(Random Samples Consensus,RANSAC)算法来消除粗匹配中误匹配的特征点对,最后将保留下来的精确匹配角点通过透视变换模型对图像进行配准。实验表明,该方法在光照、平移和模糊、旋转和尺度、视角变化具有更优的性能,提高了SURF算法的配准精度。     

Robust similarity registration technique for volumetric shapes represented by characteristic functions

This paper proposes a novel similarity registration technique for volumetric shapes implicitly represented by characteristic functions (CFs). Here, the calculation of rotation parameters is considered as a spherical cross-correlation problem and the solution is therefore found using the standard phase correlation technique facilitated by principal components analysis (PCA). Thus, fast Fourier transform (FFT) is employed to vastly improve efficiency and robustness. Geometric moments are then used for shape scale estimation which is independent from rotation and translation parameters. It is numerically demonstrated that our registration method is able to handle shapes with various topologies and robust to noise and initial poses. Further validation of our method is performed by registering a lung database.     

Improved Iterative Closest Point(ICP) 3D point cloud registration algorithm based on point cloud filtering and adaptive fireworks for coarse registration

ABSTRACT

Aiming at the problem of long computation time and poor registration accuracy in the current three-dimensional point cloud registration problem, this paper presents a k-dimensional Tree(KD-tree) improved ICP algorithm(KD-tree_ICP) that combines point cloud filtering and adaptive fireworks algorithms for coarse registration. On the basis of the typical KD-tree improved ICP algorithm, the point cloud filtering process and adaptive firework coarse registration process are added. Firstly, the point cloud data acquired by the 3D laser scanner is filtered. And then the adaptive fireworks algorithm is used to perform coarse registration on the filtered point cloud data. Next, the KD-tree_ICP algorithm is used to perform accurate registration on the basis of coarse registration, and the obtained translation and rotation relations are applied to the original point cloud data to obtain the result after registration. Finally, 3D point clouds of physical models of five statues are used for experimental verification, including error analysis, stability analysis and comparison with other algorithms. The experimental results show that the method proposed in this paper has greatly improved the calculation speed and accuracy, and the algorithm is stable and reliable, which can also be applied to the reconstruction of 3D building models, restoration of cultural relics, precision machining and other fields.     

The angular difference function and its application to image registration

The estimation of large motions without prior knowledge is an important problem in image registration. In this paper, we present the angular difference function (ADF) and demonstrate its applicability to rotation estimation. The ADF of two functions is defined as the integral of their spectral difference along the radial direction. It is efficiently computed using the pseudopolar Fourier transform, which computes the discrete Fourier transform of an image on a near spherical grid. Unlike other Fourier-based registration schemes, the suggested approach does not require any interpolation. Thus, it is more accurate and significantly faster.     

基于相位相关的亚像素图像配准

提出一种基于相位相关的图像对准方法,该方法精确度达到亚像素级别。相位相关是一种建立在傅立叶平移理论上的图像对准方法,通过计算两幅图像的傅里叶交叉能量谱得到两幅图像的位置差,对图像进行对准。将使用曲线拟合的思想对交叉能量谱的基础上进行进一步的运算,从而将相位相关方法的计算精度推广到亚像素级别。     

3D registration of human face using evolutionary computation and Kriging interpolation

This paper proposes a fast and robust 3D human face geometric data registration strategy dedicated for image-guided medical applications. The registration scheme is composed of a coarse transformation stage and a fine-tuning stage. In the first stage, fuzzy c-mean is used to reduce the data amount of template 3D image, and evolutionary computation is implemented to find optimal initial pose for the Iterative Closest Point plus k-dimensional (KD) tree scheme. In the second stage, the huge reference image data are replaced by a Kriging model. The time-consuming search for corresponding points in evaluating the degree of misalignment is substituted by projecting the points in the template image onto the model. To illustrate the validity and applicability of the proposed approach, a problem composed of 174 635 points reference image and an 11 280 points template image is demonstrated. Computational results show that our approach accelerates the registration process from 1361.28 seconds to 432.85 seconds when compared with the conventional ICP plus K-D tree scheme, while the average misalignment reduces from 11.35 mm to 2.33 mm. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008