一种基于近似点光源模型的光度立体视觉系统 标定方法

文章研究了一种由近似点光源组成的光度立体视觉系统的标定方法。由于光源自身的非均匀发光特性,传统的平行光和点光源模型不再适用,为实现精确的三维法向重建,必须对光源及光照场进行精确的标定。论文首先根据光源自身的发光特性建立其辐射度模型,进而提出了一种两步标定策略,对光源位置参数及主光轴参数进行标定。以相机作为参照系,提出了一种多球标定方法,用于计算各个光源的位置参数,该方法利用相机观察到的多个球面高光点图像坐标,结合球体轮廓线及球体半径实现了光源位置的精确计算。为实现光源主光轴参数的估算,提出了一种基于参考平面的标定方法,通过分析近似点光源照射下的平面亮度分布,结合等亮度线拟合策略实现了光源主光轴方向参数的计算。基于系统标定参数以及光源的发光模型,就可以对每个场景点的入射光条件进行精确建模,从而实现精确的法向计算。在实验部分,分别对标准几何物体和自由曲面物体进行三维重建实验,并与传统的平行光及点光源模型进行了对比。结果显示,所提出的标定方法能够在非均匀光照条件下获得精确的三维法向重建结果。     

Photometric stereo using LCD displays

This paper considers the problem of shape-from-shading using nearby extended light sources. The paper reviews a number of methods that employ nearby illuminants, and describes a new technique that assumes a rectangular planar nearby distributed uniform isotropic illuminant. It is shown that such a light source illuminating a small Lambertian surface patch is equivalent to a single isotropic point light source at infinity, in the absence of shadowing. A closed-form solution is given for the equivalent point light source direction in terms of the illuminant corner locations. Equivalent point light sources can be obtained for distinct illuminant patterns allowing standard photometric stereo algorithms to be used. An extension is given to the case of a rectangular planar illuminant with arbitrary radiance distribution. Experimental results are shown demonstrating the application of the theory to photometric stereo using illumination from a LCD computer monitor. Details on the photometric calibration of the illumination source and image acquisition device are provided.     

Determining surface orientations of specular surfaces by using the photometric stereo method

The orientation of patches on the surface of an object can be determined from multiple images taken with different illumination, but from the same viewing position. The method, referred to as photometric stereo, can be implemented using table lookup based on numerical inversion of reflectance maps. Here we concentrate on objects with specularly reflecting surfaces, since these are of importance in industrial applications. Previous methods, intended for diffusely reflecting surfaces, employed point source illumination, which is quite unsuitable in this case. Instead, we use a distributed light source obtained by uneven illumination of a diffusely reflecting planar surface. Experimental results are shown to verify analytic expressions obtained for a method employing three light source distributions.     

双光源环境下非真实感绘制色调获取方法研究

非真实感绘制场景中客体明暗色调的合理性和协调性直接影响到最后生成画面的效果。对双光源环境下非真实感绘制场景中客体色调的获取方法进行研究,提出一种双光源下基于“小球法”的色调信息提取方法,该方法假设场景中的客体表面均匀覆盖了一层大小相同的小球,相邻的两个小球间保持近似相切状态,通过计算每一个小球在双光源状态下被其它小球遮档的程度来确定该小球中心位置的明暗信息,以及每一个小球在双光源环境下所在位置的近似镜面反射光亮度。在绘制阶段,可利用这些色调信息控制在相应位置绘制“笔触”的色调,从而保证所生成的非真实感画面场景整体色调的合理性。该方法可应用于双光源下各种非真实感绘制明暗信息的获取,并且具有结构简单、计算速度快和可方便实现对场景中局部信息的调整等特点。实验结果验证了提出的方法的有效性。     

图像亮度线索下的单目深度信息提取

目的 深度信息的获取是3维重建、虚拟现实等应用的关键技术,基于单目视觉的深度信息获取是非接触式3维测量技术中成本最低、也是技术难度最大的手段。传统的单目方法多基于线性透视、纹理梯度、运动视差、聚焦散焦等深度线索来对深度信息进行求取,计算量大,对相机精度要求高,应用场景受限,本文基于固定光强的点光源在场景中的移动所带来的物体表面亮度的变化,提出一种简单快捷的单目深度提取方法。方法 首先根据体表面反射模型,得到光源照射下的物体表面的辐亮度,然后结合光度立体学推导物体表面辐亮度与摄像机图像亮度之间的关系,在得到此关系式后,设计实验,依据点光源移动所带来的图像亮度的变化对深度信息进行求解。结果 该算法在简单场景和一些日常场景下均取得了较好的恢复效果,深度估计值与实际深度值之间的误差小于10%。结论 本文方法通过光源移动带来的图像亮度变化估计深度信息,避免了复杂的相机标定过程,计算复杂度小,是一种全新的场景深度信息获取方法。     

基于平面镜的摄像机内参数线性标定方法

提出了一种全新的基于平面镜的摄像机标定方法。该方法无需任何标定物,利用平面镜的反射性质和灭点、灭线等理论,通过移动摄像机或平面镜作至少2次平移运动(各次平移间应存在旋转运动)即可线性地标定摄像机的内参数。模拟图像实验和真实图像实验表明所提出的方法能快速、方便地对摄像机进行标定,具有较好的稳定性和精度。     

利用双目立体视觉校准铣刀位置的方法研究

结合工业现场的需要,提出了一种基于双目立体视觉的快速测量铣刀盘刀具位置的方法。系统可在现场进行标定,并详细地讲述了该系统的标定原理和标定方法;标定系统时,仅需要一块二维标定板,不需要特殊的附加光源,结构简单,速度快,标定精度高;该测量方法克服了传统方法的需要接触测量、速度慢等缺点,能够实时进行自动快速测量,为工业在线生产提出了广泛地应用前景;此方法也可以应用于对工件尺寸的在线检测、孔洞位置的在线检测等。     

On planar path transformation

A flip or edge-replacement is considered as a transformation by which one edge e of a geometric object is removed and an edge f (f≠e) is inserted such that the resulting object belongs to the same class as the original object. Here, we consider Hamiltonian planar paths as geometric objects. A technique is presented for transforming a given planar path into another one for a set S of n points in convex position in the plane. Under these conditions, we show that any planar path can be transformed into another planar path by at most 2n−5 flips. For the case when the points are in general position we provide experimental results regarding transformability of any planar path into another. We show that for n?8 points in general position any two paths can be transformed into each other. For n points in convex position we show that there are n²n−2 directed Hamiltonian planar paths. An algorithm is presented which uses flips of size 1 and flips of size 2 to generate all such paths with O(n) time between the generation of two successive paths.     

Calibration of mirror position and extrinsic parameters in axial non-central catadioptric systems

We propose a novel calibration method for catadioptric systems made up of an axial symmetrical mirror and a pinhole camera with its optical center located at the mirror axis. The calibration estimates the relative camera/mirror position and the extrinsic rotation and translation w.r.t. the world frame. The procedure requires a single image of a (possibly planar) calibration object. We show how most of the calibration parameters can be estimated using linear methods (Direct-Linear-Transformation algorithm) and cross-ratio. Two remaining parameters are obtained by using non-linear optimization. We present experimental results on simulated and real images.     

Efficient Parallel Algorithms for Planar st -Graphs

   Abstract. Planar st -graphs find applications in a number of areas. In this paper we present efficient parallel algorithms for solving several fundamental problems on planar st -graphs. The problems we consider include all-pairs shortest paths in weighted planar st -graphs, single-source shortest paths in weighted planar layered digraphs (which can be reduced to single-source shortest paths in certain special planar st -graphs), and depth-first search in planar st -graphs. Our parallel shortest path techniques exploit the specific geometric and graphic structures of planar st -graphs, and involve schemes for partitioning planar st -graphs into subgraphs in a way that ensures that the resulting path length matrices have a monotonicity property [1], [2]. The parallel algorithms we obtain are a considerable improvement over the previously best known solutions (when they are applied to these st -graph problems), and are in fact relatively simple. The parallel computational models we use are the CREW PRAM and EREW PRAM.     

Single view compositing with shadows

In this paper, we describe how geometrically correct and visually realistic shadows may be computed for objects composited into a single view of a target scene. Compared to traditional single view compositing methods, which either do not deal with the shadow effects or manually create the shadows for the composited objects, our approach efficiently utilizes the geometric and photometric constraints extracted from a single target image to synthesize the shadows consistent with the overall target scene for the inserted objects. In particular, we explore (i) the constraints provided by imaged scene structure, e.g. vanishing points of orthogonal directions, for camera calibration and thus explicit determination of the locations of the camera and the light source; (ii) the relatively weaker geometric constraint, the planar homology, that models the imaged shadow relations when explicit camera calibration is not possible; and (iii) the photometric constraints that are required to match the color characteristics of the synthesized shadows with those of the original scene. For each constraint, we demonstrate the working examples followed by our observations. To show the accuracy and the applications of the proposed method, we present the results for a variety of target scenes, including footage from commercial Hollywood movies and 3D video games.     

Estimation of mirror shape and extrinsic parameters in axial non-central catadioptric systems

We propose a method to estimate the mirror shape and position, and the extrinsic parameters in axial non-central catadioptric systems (i.e. systems with an axial symmetrical mirror and a pinhole camera with its optical center located at the mirror axis). Our method requires one or more images of a planar calibration pattern consisting of points and lines with known position (e.g. a checkerboard), and that the camera be internally calibrated. We also present an alternative algorithm for the particular case of catadioptric systems with spherical mirror, were the estimation is achieved by fitting quartic curves to the images of lines of the calibration pattern. An analytical solution is presented for every method. Each analytical solution is then refined by a non-linear optimization procedure. We present experimental results, on simulated and real images, that demonstrate the validity of our work.     

Feasibility of a specialized ground light source for night-time low-light calibration

The day/night band (DNB) of the Visible Infrared Imaging Radiometer Suite can not only identify surface characteristics when illuminated by moonlight, but can also detect night-time radiation from ground active light sources. Accordingly, the low-light sensor can be calibrated by comparing upward active radiation with actual observed digital counts, which has attracted new research interest. In this article, the first attempt to calibrate the high-gain stage of the DNB using a specialized ground light source was introduced. The DNB imaging rule of the target pixel at various observation geometries was analysed based on the long-term monitoring of a light source set in Dunhuang, Gansu Province, China. The radiance caused by emissive radiation varied greatly over time in a 16-day cycle. To solve the major problem of the Dunhuang light source when used for low-light calibration, a light source that emits homogenous radiation intensity in most upward directions was custom designed. Based on this device, a comprehensive low-light calibration scheme was proposed and three verification experiments were performed in Mingguang, Anhui Province, China. The predicted in-band DNB at-aperture radiance values of the 3-day experiments were 3.86 × 10^?9, 4.38 × 10^?9, and 5.27 × 10^?9 W cm^?2 sr^?1, respectively, while the actual observed DNB radiance values were 4.53 × 10^?9, 5.06 × 10^?9, and 5.57 × 10^?9 W cm^?2 sr^?1, respectively. With deviations of 14.8%, 13.4%, and 5.4%, respectively, the calibration result based on the specialized light source was generally in good agreement with the operational calibration result, and thus the feasibility of the device for low-light calibration had been preliminarily verified.     

双目系统全视域采样的SVM标定模型分析

目的 为改善摄像机间接标定采样不全、模型表达模糊问题,实现小视场下检测视域完备采样,提出一种基于双目系统全视域采样的支持向量机（SVM）标定方法。方法 该方法利用六角晶格标定板靶点序号可读特点为基础,采集整个双目系统有效视域中检测点的视差坐标、世界坐标并建立完备的样本集。选取SVM对该样本集进行训练,将SVM算法得到的模型参数代入其决策函数中进行求解,获得公式化的标定模型。由于六角晶格标定板的四角和中心分布了5个互为非中心对称的多边形,可在标定板部分区域被采集的情况下获取标定板位姿信息,进而读取采集的各靶点序号。通过上下移动标定板,利用HALCON算子获取图像中各靶点的序号,建立双目视觉系统检测区域的完备样本集。最后,利用SVM算法训练样本获得标定模型,可以明确表达出标定模型的数学形式。结果 与传统采样建立的模型进行对比分析,实验结果表明该方法建立模型的标定误差减小了24.51%,降低了标定模型在传统方法未采样区域的标定误差,证明了该方法的可行性。结论 提出一种基于双目系统全视域采样的支持向量机标定方法,通过非中心对称的多边形确定标定板上靶点的序号,实现双目视觉系统检测视域的完备采样。实验结果表明该方法提高了摄像机间接标定的精度,具有良好的适用性和鲁棒性,适用于小视域内双目视觉系统的间接标定。     

使用人眼几何特征的视线追踪方法

目的 视线追踪是人机交互的辅助系统,针对传统的虹膜定位方法误判率高且耗时较长的问题,本文提出了一种基于人眼几何特征的视线追踪方法,以提高在2维环境下视线追踪的准确率。方法 首先通过人脸定位算法定位人脸位置,使用人脸特征点检测的特征点定位眼角点位置,通过眼角点计算出人眼的位置。直接使用虹膜中心定位算法的耗时较长,为了使虹膜中心定位的速度加快,先利用虹膜图片建立虹膜模板,然后利用虹膜模板检测出虹膜区域的位置,通过虹膜中心精定位算法定位虹膜中心的位置,最后提取出眼角点、虹膜中心点等信息,对点中包含的角度信息、距离信息进行提取,组合成眼动向量特征。使用神经网络模型进行分类,建立注视点映射关系,实现视线的追踪。通过图像的预处理对图像进行增强,之后提取到了相对的虹膜中心。提取到需要的特征点,建立相对稳定的几何特征代表眼动特征。结果 在普通的实验光照环境中,头部姿态固定的情况下,识别率最高达到98.9%,平均识别率达到95.74%。而当头部姿态在限制区域内发生变化时,仍能保持较高的识别率,平均识别率达到了90%以上。通过实验分析发现,在头部变化的限制区域内,本文方法具有良好的鲁棒性。结论 本文提出使用模板匹配与虹膜精定位相结合的方法来快速定位虹膜中心,利用神经网络来对视线落点进行映射,计算视线落点区域,实验证明本文方法具有较高的精度。     

Unstructured Light Scanning Robust to Indirect Illumination and Depth Discontinuities

Reconstruction from structured light can be greatly affected by indirect illumination such as interreflections between surfaces in the scene and sub-surface scattering. This paper introduces band-pass white noise patterns designed specifically to reduce the effects of indirect illumination, and still be robust to standard challenges in scanning systems such as scene depth discontinuities, defocus and low camera-projector pixel ratio. While this approach uses unstructured light patterns that increase the number of required projected images, it is up to our knowledge the first method that is able to recover scene disparities in the presence of both indirect illumination and scene discontinuities. Furthermore, the method does not require calibration (geometric nor photometric) or post-processing such as phase unwrapping or interpolation from sparse correspondences. We show results for a few challenging scenes and compare them to correspondences obtained with the Phase-shift method and the recently introduced method by Gupta et al., designed specifically to handle indirect illumination.     

均匀线阵中幅相及位置误差的快速校正方法

着重研究了针对均匀线阵中由各阵元幅度相位不一致性及位置误差的综合影响引起的阵列流形误差的校正问题。该方法利用单信号源(可以为事先设置的校正源或某目标源)，无须准确知道信号源的波迭方向，只须在校正过程中将阵列天线以已知角度旋转两次．即可对各阵元的幅度、相位及位置因子作较精确的估计，从而估计出综合误差存在情况下的阵列流形，并可同时估计信号源的波达方向。该方法无需迭代．计算简单快速．且具有较高的估计精度。计算机模拟实验结果表明了本文方法的有效性。     

Three-dimensional position estimation method via AM pulse light modulation and an application to control multiple UAVs

ABSTRACT

In this study, a three-dimensional position estimation is proposed to estimate self-position via amplitude-modulated pulse infrared light to solve the problem of position estimation from an existing device used in the cooperation of unmanned air vehicles (UAVs) that can be used in various environments. The proposed method estimates self-position as the distance between the receiver and the transmitter and the angle of direction by attenuation of amplitude-modulated pulse infrared light. By attaching a transmitter to a leader UAV and a receiver to follower UAVs, the cooperation flight can be achieved. The following three experiments are conducted: (1) evaluation of the accuracy of self-position estimation, (2) implementation of the proposed method on an UAV, and (3) evaluation of the robustness of the proposed method in an outdoor environment. The findings confirm the effectiveness of the proposed method. The proposed approach allows estimation of position with high speed, precision in various environments, and low-cost devices.     

基于平面格网的非量测CCD相机自标定

传统的相机标定方法通常需要建立复杂3维标定块或高精度3维控制场,在实际应用中受到了一定的限制。本文采用平面控制格网作为标定块,根据相机的理想模型确定内方位元素,利用2维直接线性变换和共线方程分解出相机的外方位元素初值,采用改进的Hough变换算法检测标定图像中的格网直线并利用最小二乘法拟合出最佳直线,通过求直线的交点得到标定格网点的像坐标。最后利用自检校光线束法平差进行相机的精确标定。实际图像数据实验结果表明,主点和焦距的标定精度分别达到了0.2像素和0.3像素左右。可以满足高精度近景3维量测的要求。     

变形镜驱动器参数自动标定系统的设计

变形镜驱动器各通道的offset参数和gain参数直接影响到变形镜驱动器输出电压的精度,采用人工方式对其进行标定,不仅效率低,费时费力,而且参数的一致性也无法得到保证。为此,首先设计了AD采集器,其内部仅含一片24bit的高精度AD采集芯片,其并行的16端子输入接口通过继电器与采集芯片相连,内含的微处理器负责完成输入通道切换、采集时序控制和采集数据上传等功能;同时,还设计了上位机应用程序,实现了对变形镜驱动器以及AD采集器通信链路检查、标定过程控制参数设置、采集通道切换以及数据处理等相关操作命令和状态数据的收发,使得变形镜驱动器、AD采集器和上位机组成一套驱动器参数自动标定系统。结果表明,该系统不仅使得一套变形镜驱动器参数标定时间由原来的两个工作日缩短到2小时左右,而且提高了参数标定精度和一致性,其96通道零电压的最大误差由18.2mV降为13.5mV,均方差由原来的6.2mV降低为4.3mV,其输出最高电压120.000V的最大误差由23.4mV降为17.5mV,均方差由原来的10.1mV降低为8.3mV。