双目系统中图像随机噪声的分析

图像立体匹配的精度是保证双目系统测量准确性的重要因素，也是系统研制过程中需要解决的难点问题。对影响双目系统中图像立体匹配精度的随机噪声建立了一种局部统计量符合高斯分布的简化统计模型，基于这个噪声模型设计了滤波方案。实验表明，滤波后图像立体体匹配的精度优于1／4像素。     

Estimating three-dimensional vehicle motion in an outdoor scene using stereo image sequences

We propose a motion estimation system that uses stereo image pairs as the input data. To perform experimental work, we also obtain a sequence of outdoor stereo images taken by two metric cameras. The system consists of four main stages, which are (1) determination of point correspondences on the stereo images, (2) correction of distortions in image coordinates, (3) derivation of 3D point coordinates from 2D correspondences, and (4) estimation of motion parameters based on 3D point correspondences. For the first stage of the system, we use a four-way matching algorithm to obtain matched point on two stereo image pairs at two consecutive time instants (t_i and t_{i + 1}). Since the input data are stereo images taken by cameras, it has two types of distortions, which are (i) film distortion and (ii) lens distortion. These two distortions must be corrected before any process can be applied on the matched points. To accomplish this goal, we use (i) bilinear transform for film distortion correction and (ii) lens formulas for lens distortion correction. After correcting the distortions, the results are 2D coordinates of each matched point that can be used to derive 3D coordinates. However, due to data noise, the calculated 3D coordinates to not usually represent a consistent rigid structure that is suitable for motion estimation; therefore, we suggest a procedure to select good 3D point sets as the input for motion estimation. The procedure exploits two constraints, rigidity between different time instants and uniform point distribution across the object on the image. For the last stage, we use an algorithm to estimate the motion parameters. We also wish to know what is the effect of quantization error on the estimated results; therefore an error analysis based on quantization error is performed on the estimated motion parameters. In order to test our system, eight sets of stereo image pairs are extracted from an outdoor stereo image sequence and used as the input data. The experimental results indicate that the proposed system does provide reasonable estimated motion parameters.     

成像降质效应对前视红外图像质量的影响分析

前视红外图像质量对于目标检测与跟踪算法能否准确定位目标具有十分重要的作用。目前对自动目标识别领域前视红外图像质量的研究大都与场景内容相关,而忽视了成像降质效应对图像质量的影响。本文通过分析噪声和模糊两种降质效应的成因,结合前人的模拟方法,得到了高性能前视红外成像设备获取的图像中噪声和模糊模拟的新方法。将该方法用于分析不同程度噪声和模糊对图像质量的影响,结果显示,噪声对于图像质量有害,而模糊对于图像质量并不一定有害。基于这一结论,采用去噪方法对前视红外图像进行预处理。对比实验结果表明,该方法比去模糊方法能更有效地提高匹配概率。     

Three-dimensional tracking of multiple skin-colored regions by a moving stereoscopic system

A system that performs three-dimensional (3D) tracking of multiple skin-colored regions (SCRs) in images acquired by a calibrated, possibly moving stereoscopic rig is described. The system consists of a collection of techniques that permit the modeling and detection of SCRs, the determination of their temporal association in monocular image sequences, the establishment of their correspondence between stereo images, and the extraction of their 3D positions in a world-centered coordinate system. The development of these techniques has been motivated by the need for robust, near-real-time tracking performance. SCRs are detected by use of a Bayesian classifier that is trained with the aid of a novel technique. More specifically, the classifier is bootstrapped with a small set of training data. Then, as new images are being processed, an iterative training procedure is employed to refine the classifier. Furthermore, a technique is proposed to enable the classifier to cope with changes in illumination. Tracking of SCRs in time as well as matching of SCRs in the images of the employed stereo rig is performed through computationally inexpensive and robust techniques. One of the main characteristics of the skin-colored region tracker (SCRT) instrument is its ability to report the 3D positions of SCRs in a world-centered coordinate system by employing a possibly moving stereo rig with independently verging CCD cameras. The system operates on images of dimensions 640 x 480 pixels at a rate of 13 Hz on a conventional Pentium 4 processor at 1.8 GHz. Representative experimental results from the application of the SCRT to image sequences are also provided.     

无人飞艇低空数码影像多视摄影测量

针对大比例尺测图的精度要求,研究低空数码影像多视摄影测量的方法.根据计算机视觉的多视几何原理建立匹配的约束条件,实现多视匹配;利用低空数码影像的重叠度大的特点,建立多基线立体,进行多片空间前方交会.不同重叠度的立体像对空间前方交会和多片空间前方交会的实验结果证明,多视摄影测量能够提高有效低空数码影像的摄影测量精度.     

立体视觉曲面重建与系统误差分析

采用立体视觉空间曲面重建技术对三维曲面表面成像进行边缘提取、图像匹配、匹配点空间位置计算等步骤,得到三维曲面表面点的空间位置,利用空间点信息对三维曲面形状进行重建,恢复曲面三维形状;并讨论了立体视觉系统的摄像机分辨率、测量范围和摄像机间距等参数之间的关系.利用该方法对堆积物表面形状及体积进行测量实验结果表明,该方法能准确、快速、方便地给出三维曲面的形状.     

基于双目相机的光栅立体印刷图像合成

目的研究一种基于双目相机的光栅立体图像合成方法。方法首先用双目相机采集场景中2个观察角度的二维平面图像数据。基于双目立体视觉理论,采用一种鲁棒性较强的基于区域分割的图像匹配方法,得到精度较高的深度图。然后分析序列视差图像的成像模型,建立一种基于双目图像对生成序列视差图像的方法,得到连续角度等间隔的序列图像。最后基于柱镜光栅的光学特性形成的立体印刷图像编码规则,对序列视差图像进行纵向条纹抽样分割,等间隔的抽取每幅序列视差图像中的对应列实现光栅立体图像的合成。结果验证了该光栅立体图像合成方法的有效性。结论基于双目相机的光栅立体合成方法,可以使立体印刷产品实现个性化、便捷化的即时输出。     

Development of a real-time stereo TEM

We have developed a real-time stereo transmission electron microscope (TEM) with tilting illumination. This system allows us to observe three-dimensional (3-D) images directly with a video-rate of 1/30 s. The system comprises two electrostatic deflectors. The first, included in the illumination system, is able to illuminate a specimen at two oblique stereoscopic angles (left and right of the optical axis) up to ±2.3 °. The second deflector, in the imaging system, is used to correct the image separation caused by defocusing of the tilted illumination. These deflectors are operated in synchronization with an NTSC video signal output from a CCD camera to record left projections on odd fields and right projections on even fields. The time series of stereo pairs is transferred to a 3-D display that enables viewing of the 3-D images without special glasses. Real-time observation of ZnO particles demonstrated that 3-D images were clear even while moving the specimen. We applied this system to observing dislocations in an Al thin film.     

结合颜色不变量和ORB特征的图像匹配算法

目的针对现有图像匹配算法因忽略图像颜色信息,使算法对光照变化的鲁棒性较差,容易造成图像误匹配的情况进行改善。方法颜色信息是区分不同图像的重要特征,通过提取彩色图像颜色不变量的方法获得图像的颜色信息,并将颜色不变量与ORB特征描述子相结合,形成一种C-ORB算法,C-ORB算法在保留ORB算法优点的基础上,能够将图像的色彩信息应用到图像匹配过程当中。结果实验结果表明,结合了颜色不变量的ORB算法在保留了ORB算法运算速度快的基础上,能够在不同光照强度、不同光照方向下,获得更多的图像细节,匹配更为准确。结论在移动手机、平板电脑等移动终端设备进行印刷品识别与匹配时,能够在保证识别图像识别速度的基础上进一步减少图像的误匹配。     

Optical Filtering Approach to Regularized Tracking of an Object's Position and Orientation

We present a regularized nonlinear least-squares algorithm for tracking the position and the orientation of a known object by using an active camera and an optical correlator situated between digital preprocessing and postprocessing operations. The numerical minimization required by the regularized least-squares solution is implemented by use of a rapid look-up table method. Performance of the algorithm is evaluated through a Monte Carlo sensitivity analysis that incorporates models for lens blur, image noise, illumination variation, and partial occlusion. This analysis shows robust performance with respect to image noise, partial occlusion of the object, and errors in the camera pan and tilt used to follow the moving object. The limiting factors in the algorithm's performance are errors in the preprocessing step used to scale and rotate the input video images. These errors should be maintained within 6% and 3 degrees , respectively.     

扩展核线模型在线阵CCD卫星遥感影像立体匹配中的应用

核线(epipolarity)是立体摄影测量分析像对几何关系的一个基本概念,在摄影测量的处理当中有着重要的应用价值,但经典的核线关系仅适用于框幅式中心投影影像.针对线阵CCD卫星遥感立体影像,将基于投影轨迹法的扩展核线模型引用到影像匹配中, 提出了一种基于动态核线的近似直线约束方法,实现了线阵CCD立体影像匹配的核线约束;在此核线模型应用的基础上,提出了一种基于核线的成像约束方法.试验结果表明前一种方法在匹配速度方面比传统方法有显著的提高,后一种方法在匹配精度、可靠性方面比传统方法有明显提高.     

基于局部纹理特性和图像分割的分步立体匹配

为解决立体匹配难以兼顾精度和速度的问题,提出一种分步立体匹配方法。根据局部纹理特性对图像对进行灰度变换;然后利用均值漂移算法分割参考图像,以任意大小和形状的分割区域为支持窗口进行初次匹配,形成基于色彩分割的视差约束;再以固定窗口为支持窗口进行二次匹配,获得初始视差图;最后通过可信度分类优化初始视差。实验结果表明所提出的算法兼有较高的匹配速度和精度。     

基于超拉普拉斯先验与核谱特性噪声图像盲去模糊

现有大部分盲图像去模糊方法对噪声敏感,即使少量的噪声可大大降低恢复图像的质量．考虑到模糊图像中同时隐含有清晰图像信息和模糊核信息,我们同时利用卷积核谱特性先验和清晰图像梯度域超拉普拉斯先验联合建立含噪图像盲去模糊模型,较单独使用卷积核先验与清晰图像先验建模更合理,也能获得更精确的估计图像．本文借助于Hessian矩阵,利用模糊图像及卷积核联合生成先验子,而非单独的估计图像先验子,建立优化模型．求解模型时,通过迭代策略交替细化模糊核和清晰图像．在清晰图像恢复阶段,因存在超拉普拉斯先验项,提出用变量分离法计算清晰图像．清晰图像采用快速傅里叶变换及封闭阈值公式求解,以提高优化速度．实验结果表明：与其他方法相比,本文方法能获得更鲁棒的模糊核和更精确的清晰图像,且收敛速度更快．     

Image speckle noise denoising by a multi-layer fusion enhancement method based on block matching and 3D filtering

In order to improve speckle noise denoising of block matching and 3D filtering (BM3D) method, an image frequency-domain multi-layer fusion enhancement method (MLFE-BM3D) based on nonsubsampled contourlet transform (NSCT) has been proposed. The method designs an NSCT hard threshold denoising enhancement to preprocess the image, then uses fusion enhancement in NSCT domain to fuse the preliminary estimation results of images before and after the NSCT hard threshold denoising, finally, BM3D denoising is carried out with the fused image to obtain the final denoising result. Experiments on natural images and medical ultrasound images show that MLFE-BM3D method can achieve better visual effects than BM3D method, the peak signal to noise ratio (PSNR) of the denoised image is increased by 0.5?dB. The MLFE-BM3D method can improve the denoising effect of speckle noise in the texture region, and still maintain a good denoising effect in the smooth region of the image.     

基于GA-BP算法噪声检测的椒盐噪声滤波算法

基于噪声检测的中值滤波器已广泛用于消除图像中的椒盐噪声,然而在高噪声密度情况下,对噪声像素的定位不准确很容易造成图像边缘的模糊.本文提出了一种基于GA-BP的椒盐噪声滤波算法,克服了这一缺陷.算法首先用遗传算法优化的BP网络对图像中的噪声像素定位,然后引入保边函数和PRP算法求目标函数的极值进而实现图像的去噪处理.实验...     

运动模糊不变量研究

图像的运动模糊是摄像机在曝光期间与场景间相对运动而造成的图像退化,但现有算法对运动模糊的恢复精度有限.本文分析了图像运动模糊的退化模型,提出了运动模糊不变量的概念,证明了图像零阶矩、第一个Hu不变矩和SIFT为运动模糊不变量,并给出了测试和应用.结果表明运动模糊不变量是存在的,而且可以用于不确定性运动模糊图像的匹配和识别领域.     

多尺度空间基于集中度判定的二维稳像算法

二轴运动平台由于震动、非一致性摩擦等因素造成视频序列出现抖动,同时外部噪声,光照变化等对图像特征提取、匹配等存在严重影响导致运动向量估计错误,本文提出一种在多尺度空间基于集中度判定的二维稳像算法.通过在多尺度空间下提取视频序列的不变特征,采用深度优先最邻近搜索算法,寻找匹配点对,然后计算匹配点对的集中度,通过最优集中化...     

3D photon counting integral imaging with unknown sensor positions

Photon counting techniques have been introduced with integral imaging for three-dimensional (3D) imaging applications. The previous reports in this area assumed a priori knowledge of exact sensor positions for 3D image reconstruction, which may be difficult to satisfy in certain applications. In this paper, we extend the photon counting 3D imaging system to situations where sensor positions are unknown. To estimate sensor positions in photon counting integral imaging, scene details of photon counting images are needed for image correspondences matching. Therefore, an iterative method based on the total variation maximum a posteriori expectation maximization (MAP-EM) algorithm is used to restore photon counting images. Experimental results are presented to show the feasibility of the method. To the best of our knowledge, this is the first report on 3D photon counting integral imaging with unknown sensor positions.     

Specularity and shadow detection for the multisource photometric reconstruction of a textured surface

Textured surface analysis is essential for many applications. In this paper, we present a three-dimensional (3D) recovery approach for real textured surfaces based on photometric stereo. The aim is to be able to reconstruct the textured surfaces in 3D with a high degree of accuracy. For this, the proposed method uses a sequence of six images and a Lambertian bidirectional reflectance distribution function (BRDF) to recover the surface height map. A hierarchical selection of these images is employed to eliminate the effects of shadows and highlights for all surface facets. To evaluate the performances of our method, we compare it to other traditional photometric stereo methods on real textured surfaces using six or more images.     

Real-time pedestrian classification exploiting 2D and 3D information

A new approach for standing and walking pedestrian detection using pattern matching and exploiting both 2D image information and 3D dense stereo information is proposed. Because 3D information accuracy does not allow the direct classification of the 3D shape, a combined 3D-2D method is proposed. The 3D data are used in an innovative way for pedestrian hypotheses generation, scale and depth estimation and 2D models selection. Also the 3D hypotheses allow the corresponding 2D image region of interest selection and the 2D hypothesis generation. The 2D hypothesis consists of the object-s external edges obtained by an edge extraction and a depth coherency-based filtering out process. The scaled models are matched against the selected hypothesis using an elastic high-speed matching based on the Chamfer distance. The method has been tested on synthetic and real-world scenarios.