基于改进SIFT-ICP算法的物体点云建模方法

为了实现对任意摆放的物体的识别与位姿估计,需要得到物体的三维模型,提出了一种高效准确的物体三维点云模型构建方法。首先通过RGB-D传感器获取物体多角度数据,然后利用尺度不变特征变换(SIFT)特征点匹配和改进的迭代最近点(ICP)算法计算出各角度下传感器的相对位姿,进而生成目标物体所在场景的完整点云,并通过物体分割和点云后处理得到目标物体的三维点云模型。实验结果显示,构建的物体点云模型清晰且不失真,并保留了表面完整的特征信息。     

光笔视觉三维测量中光斑图像点立体匹配方法

针对光笔视觉三维测量中左右两幅图像中光斑图像点的立体匹配难题,引入SoftPosit 算法将其转换为二维像点与三维物点对应关系未知情况下的位姿参数估计问题.通过将对应匹配和位姿估计融合在一起,进行捆绑迭代寻优,分别完成左右两幅图像中光斑图像点与空间光笔光斑的对应匹配,从而实现了左右两幅图像中对应光斑图像点的立体匹配.该方法基于模型,与空间点分布模式、点的数量及图像灰度无关,实验结果表明该方法切实可行,也可用于特征点的识别.     

基于脊线预配准的肋骨骨折钢板预弯算法研究

提出了一种可以在肋骨骨折手术前获取目标固定钢板几何参数的仿真方法.其主要操作步骤包括:通过CT扫描获取骨折部位的序列断层图像,利用Curvelet变换对CT序列图像进行去噪处理并建立三维模型;根据肋骨的形状特征,提取两段断骨的脊线进行空间预配准;并驱动两段断骨进行模型预配准.使用网格模型内三角面片的面积特征分割得到两个...     

利用离散平稳小波变换改进NURBS二次曲面拟合方法

为了提高二次曲面的拟合精度,提出了一种基于离散平稳小波变换的NURBS二次曲面的拟合方法。 首先利用格网化方法得到二次曲面模型点云数据的高程图像及其高程序列,其次对此高程序列进行离散平稳小波变换提取出二次曲面模型表面的特征点,最后利用提取的特征点实现高精度NURBS二次曲面拟合。实验结果表明,该方法与NURBS拟合方法相比,球面和圆锥面拟合结果的均方根误差分别降低了55.79%和50.47%,具有较高的拟合精度。     

局部特征点的鲁棒性数字稳像

针对视频序列受载体的影响,本文提出了基于Harris局部特征点的数字稳像方法。局部特征点在图像处理、模式识别、计算机视觉以及目标检测与跟踪领域得到了广泛的研究。首先,利用Harris算法提取动载体摄像系统中每帧图像的局部特征点,采用基于归一化互相关匹配算法进行局部特征点的匹配,提出了一种双向搜索的匹配策略来提取具有较强鲁棒性的特征点;然后,利用随机抽样一致算法进一步筛选所选定的特征点,并将保留下来的精确匹配特征点带入仿射运动模型求出全局运动矢量;最后,提取全局运动矢量中的抖动参数,并对原始图像进行补偿。实验结果表明,该方法可以快速有效地处理图像序列的仿射运动,水平和垂直方向的精确度小于1 pixel,并且具有较好的鲁棒性,可用于动载体摄像系统。     

单视旋转显微镜三维测量技术研究

为获得微小器件的三维形貌,该文利用单视旋转显微镜通过旋转被测物体或成像系统采集被测物体的多角度图像序列,应用多视三维重建算法处理序列图像。通过改进宏观视觉领域基于面片的立体视觉方法,实现单目旋转视频显微镜的被测工件三维重建技术;并根据物体相对于相机倾斜角度的范围来细化扩展半径,增加原算法中的扩展方向,针对扩展面片法向量相对于相机角度的不同,设置不同扩展半径和扩展次数,解决重建过程中物体90°垂直侧面重建面片成条状和稀疏的问题。实验结果表明:通过改进算法,相同的显微镜图像序列进行三维重建的点个数增加2.2倍;同时对于物体侧面等垂直表面处,重建表面的漏洞和分层问题有明显的改善,最小相对测量误差达0.44%。     

基于多视角区域生长的复杂点云模型分割

为提高三维点云模型在特征模糊区域的分割精度,提出了一种借助多视角区域生长的分割方法。基于网格法向量方向相异性原则,初次将模型划分为不同类别的子区域,在相应区域建立点云与多视角距离图像的一一映射关系。利用Canny算子对灰度的敏锐性获取独立连通域并计算其重心坐标,根据对应关系在三维点云中提取对应点作为种子点,然后引入网格法向量的偏移角度分离邻接面,同时对剩余彼此独立的分割面按照迭代搜索最近点的原则进行提取,并利用KNN算法去除离群点实现分割优化。在选取的模型数据集上进行实验,结果表明该方法能够实现复杂点云模型的合理划分,分割精度不低于80%。     

基于几何约束的双目测距技术研究

双目测距是一种重要的机器视觉技术,常用于机器人定位与导航任务。立体匹配在双目视觉中占有重要的作用,文章基于构建的双目立体视觉系统,提出了一种基于几何约束的物体三维坐标测量方法。首先利用立体标定技术实现了对图像的畸变矫正,并对实时拍摄的左右图像进行ORB特征匹配,以检测图像中物体的特征点对。算法对估计的初始ORB特征点进行RANSAC模型估计,去除误匹配点对应。之后基于点对应之间的距离与斜率对其进行几何关系度量,利用几何约束进一步滤除匹配误差较大的点对应。实验结果说明该算法能有效降低特征的误匹配率,提高了双目测距的准确性与鲁棒性。     

基于连通性转换的三维多面体快速变形

研究了连通性转换方法,并将其应用于多面体快速变形.采用优先权控制函数控制模型特征,且可以实现点和边的同步处理.通过预先安排连通性转换提高了变形速度,通过解一个稀疏线性方程组缩短变形片的嵌入时间.之后,矫正特征点以及边界处理得到平滑变形序列.用该方法和传统变形方法分别演示了三维多面体模型的变形实例,并对实验结果和相关数据进行了分析比较.实验结果表明,该变形方法可快速实现平滑变形.     

气膜孔视觉测量系统的设计与搭建

为了重建高压涡轮导向叶片气膜孔的三维形貌特征,基于机器视觉原理,设计并搭建了非接触式的气膜孔四轴视觉测量系统。该系统主要由三坐标测量机框架、高精度回转台、叶片专用夹具和图像采集装置构成,将视觉测量技术和三坐标测量技术结合起来以应对气膜孔的检测问题。为了获取气膜孔的三维形貌数据,提出了应用景深合成技术进行孔壁三维形貌重建的方法,并通过实验进行了重建方法的验证。在实验过程中,应用该测量系统对某个高压涡轮导向叶片上的气膜孔特征进行了测量实验,获取了孔深范围内的对焦图像序列,而后对其进行了景深合成与深度信息转化,最终实现了该气膜孔的孔壁形貌的三维重建与显示,并实现了向三维点云数据的转化,充分说明了所搭建的气膜孔视觉测量系统的有效性。     

彩色伪随机编码闪光投影系统的建模及其标定

为实现仅用一幅图象在真正的三维欧氏空间重构三维场景,尤其是动态的三维场景,系统需要对彩色伪随机编码闪光投影系统进行建模和预标定。当系统参数改变后,利用伪随机编码图案的几何约束,不需重新标定,可实现自适应重构三维场景。通过对投影仪像质的的无畸变性分析,用线性模型对投影仪成像过程进行建模,并提出一种高精度的投影系统直接标定方法,通过测出靶标上各投影编码特征点在空间的三维齐次坐标,求出内外参数实现了对投影仪的预标定。标定精度达到0.1单元格,欧氏重构结果较为理想。     

Research on Multi-View Image Reconstruction Technology Based on Auto-Encoding Learning

Traditional three-dimensional (3D) image reconstruction method, which highly dependent on the environment and has poor reconstruction effect, is easy to lead to mismatch and poor real-time performance. The accuracy of feature extraction from multiple images affects the reliability and real-time performance of 3D reconstruction technology. To solve the problem, a multi-view image 3D reconstruction algorithm based on self-encoding convolutional neural network is proposed in this paper. The algorithm first extracts the feature information of multiple two-dimensional (2D) images based on scale and rotation invariance parameters of Scale-invariant feature transform (SIFT) operator. Secondly, self-encoding learning neural network is introduced into the feature refinement process to take full advantage of its feature extraction ability. Then, Fish-Net is used to replace the U-Net structure inside the self-encoding network to improve gradient propagation between U-Net structures, and Generative Adversarial Networks (GAN) loss function is used to replace mean square error (MSE) to better express image features, discarding useless features to obtain effective image features. Finally, an incremental structure from motion (SFM) algorithm is performed to calculate rotation matrix and translation vector of the camera, and the feature points are triangulated to obtain a sparse spatial point cloud, and meshlab software is used to display the results. Simulation experiments show that compared with the traditional method, the image feature extraction method proposed in this paper can significantly improve the rendering effect of 3D point cloud, with an accuracy rate of 92.5% and a reconstruction complete rate of 83.6%.     

利用 OpenGL 实现 ICT 断层图像的三维重建

基于ICT断层图像的三维重建是实现实物反求的重要途径之一．本文提出了一种基于Delaunay三角剖分思想的表面重建方法．并采用OpenGL结合VC 6．0来实现ICT图像的三维重建．最后给出了仿真结果与实际实验结果．     

Robust estimation of 3D trajectories from a monocular image sequence

This article describes a new method for object trajectory estimation that uses sequences of images taken from a monocular camera. The method integrates a Kalman filter to estimate the three‐dimensional (3D) parameters of the optical system and a lineal projective model to determine 3D point coordinates projected on the retinal plane. It works with at least three distinctive points in the image, and they are updated with correlation methods. The result is an estimation of the rotation and translation parameters between successive images within the sequence and yield to the 3D coordinates of the points selected for correspondence. The scaling problem related to 3D reconstruction is tackled via a priori information of the objects being observed. The method is tested with synthetic images to evaluate its accuracy, and later, an interesting application in autonomous navigation is presented. © 2002 Wiley Periodicals, Inc. Int J Imaging Syst Technol 12, 128–137, 2002; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10020     

AHSPP Annual Conference

Abstract

A calibration and three-dimensional (3D) reconstruction method is presented based on images reflected from planar mirrors and acquired with one camera. The geometric model of the camera–mirror set and a method of calibrating it are described. The calibration technique computes the model parameters using linear equations, and it is proved that the calibration is possible with the knowledge of only six 3D points. The reconstruction method is based on a volumetric representation. The 3D reconstruction is based on a space carving algorithm and the calibration method described in the paper. The results of the calibration and reconstruction method show the efficiency of both techniques. This set-up enables a simple and inexpensive multi-ocular system to be built to recover the 3D structure of volumes.     

双目立体视觉中特征点三维坐标重构校准研究

双目立体视觉是一种商业化较成熟的三维测量技术,左右摄像机内外参数的精确标定是实现三维重构的基础和关键。针对棋盘格和圆点两种标定板图案研究了相应的图像处理技术,实现了标定点的亚像素精度定位及其有效排序。采用基于单映性约束和非线性优化的多视角平面标定算法实现了摄像机光学及空间位置参数求解。用极线约束残差法衡量标定结果的准确度。基于桥式三坐标标准实现标定点三维重构平面度以及多平面空间夹角测量结果的校准,基于光学三坐标标准实现了标定点三维重构空间距离的校准,并分析了校准结果的不确定度。圆标志点三维重构空间距离示值误差为0.029 mm,不确定度U=24 μm（k=2）     

Three dimension reconstruction through measure-based image selection

Three dimension (3D) reconstruction is one of the research focus of computer vision and widely applied in various fields. The main steps of 3D reconstruction include image acquisition, feature point extraction and matching, camera calibration and production of dense 3D scene models. Generally, not all the input images are useful for camera calibration because some images contain similar and redundant visual information. These images can even reduce the calibration accuracy. In this paper, we propose an effective image selection method to improve the accuracy of camera calibration. Then a new 3D reconstruction algorithm is proposed by adding the image selection step to 3D reconstruction. The image selection method uses structure-from-motion algorithm to estimate the position and attitude of each camera, first. Then the contributed value to 3D reconstruction of each image is calculated. Finally, images are selected according to the contributed value of each image and their effects on the contributed values of other images. Experimental results show that our image selection algorithm can improve the accuracy of camera calibration and the 3D reconstruction algorithm proposed in this paper can get better dense 3D models than the normal algorithm without image selection.     

Fast Scene Reconstruction Based on Improved SLAM

Simultaneous location and mapping (SLAM) plays the crucial role in VR/AR application, autonomous robotics navigation, UAV remote control, etc. The traditional SLAM is not good at handle the data acquired by camera with fast movement or severe jittering, and the efficiency need to be improved. The paper proposes an improved SLAM algorithm, which mainly improves the real-time performance of classical SLAM algorithm, applies KDtree for efficient organizing feature points, and accelerates the feature points correspondence building. Moreover, the background map reconstruction thread is optimized, the SLAM parallel computation ability is increased. The color images experiments demonstrate that the improved SLAM algorithm holds better real-time performance than the classical SLAM.     

A Dual-Platform Laser Scanner for 3D Reconstruction of Dental Pieces

This paper presents a dual-platform scanner for dental reconstruction based on a three-dimensional (3D) laser-scanning method. The scanner combines translation and rotation platforms to perform a holistic scanning. A hybrid calibration method for laser scanning is proposed to improve convenience and precision. This method includes an integrative method for data collection and a hybrid algorithm for data processing. The integrative method conveniently collects a substantial number of calibrating points with a stepped gauge and a pattern for both the translation and rotation scans. The hybrid algorithm, which consists of a basic model and a compensation network, achieves strong stability with a small degree of errors. The experiments verified the hybrid calibration method and the scanner application for the measurement of dental pieces. Two typical dental pieces were measured, and the experimental results demonstrated the validity of the measurement that was performed using the dual-platform scanner. This method is effective for the 3D reconstruction of dental pieces, as well as that of objects with irregular shapes in engineering fields.     

基于稀疏激光点云数据和单帧图像融合的三维重构算法

由于激光雷达获取的深度数据非常稀疏,为了能够将深度数据与图像数据重构出稠密三维深度图,本文提出了基于稀疏激光点云数据和单帧图像融合的三维重构算法。该方法首先使用点直方图特征有效地选择对应于目标的点数据并消除体素中的非相似点;然后,使用高斯过程回归对局部深度数据建模,并通过插值获得三维深度数据,本文算法获得的三维深度点更接近基准值,并保持了目标的局部形状特征;最后,利用马尔科夫随机场对图像灰度数据和三维插值点进行融合来构建三维深度图。仿真实验结果表明:相比现有基于激光雷达数据和单目图像数据的三维重建算法,本文提出的算法将大大提升算法的鲁棒性与重构的准确度,可辅助用于复杂的城市场景中车辆的无人驾驶。