机器视觉领域的FPGA超高清图像采集系统设计

为满足当前机器视觉等领域对高清晰度成像的要求,提出了一种基于FPGA的超高清图像采集系统设计方案,通过Xilinx Artix-7系列FPGA芯片XC7A35T-CSG324搭载相关模块实现图像的采集、存储、显示。在硬件设计上选用IMX183图像传感器进行图像采集、DDR3芯片进行缓存、Flash芯片进行存储,并通过DC-DC电源模块与LDO电源模块配合使用实现多通道系统供电。软件设计上通过SPI协议进行图像传感器的配置,并对LVDS图像数据进行解析。实际测试结果表明,该系统可以实现2 000万像素的超高清图像显示,配合光学结构可应用于工业相机、监控摄像头等多种产品中,具有较高的实用价值和研究意义。     

新型三维腹腔镜系统的设计与标定（英文）

目的:为简化三维腹腔镜结构,同时保证其成像和3D视频质量,研发一种新型高清三维腹腔镜系统。创新:采用双光学通道,使用中继透镜延长图像光学传输距离,在中继透镜末端用两个相机接收内窥图像,简化腹腔镜结构,并保证了成像质量。方法:首先,根据图像传感器参数和视场需求大小确定内窥镜光学系统的有效焦距,在此基础上设计包括中继透镜在内的双通道光学镜头(图1)。其次,设计CCU用于采集和处理双通道高清视频图像,并通过HDMI输出将它们同时显示在3D显示器上,由此构建了高清三维腹腔镜系统(图4)。最后,针对双目内窥镜系统采用传统针孔模型计算距离存在较大误差的问题,通过理论分析提出一种新的距离计算方法,并在0-150 mm范围内的距离测量实验中验证该算法的有效性(图3)。结论:新型三维腹腔镜在保证成像质量的基础上简化了系统结构;采用提出的双目系统距离计算方法可以有效提高双目内窥系统的测距精度。     

基于VTK的医学图像体绘制及交互实现

由于体绘制可以保留医学图像的所有数据信息．并且能对于形状特征模糊的组织器官以及它们之间的层次关系进行三维显示．因此利用VTK开发包所提供的体绘制方法来实现CT医学图像的三维重建,并为用户提供了的交互功能,从而让医生可以多角度、多层次进行病体观察和病因的分析。     

全景三维虚拟系统构建方法研究

三维GIS系统实现了现实世界的虚拟化表达,但是所生成的三维场景缺乏真实感,全景技术的三维再现能力,虽然可弥补三维GIS表达上的缺陷,但是缺乏全局可浏览性。以环拍的图像序列为对象,利用Sift(scale invariant feature transform)算子实现全景图的自动拼接,通过Pano2VR软件实现全景图的操纵配置;结合Skyline三维软件,以福建师范大学为例,实现全景三维校园网络浏览系统。该系统具有良好沉浸性、多角度可浏览性,使用户具有真实的三维视觉体验;实现方式简单、构建快速,也为旅游景点、酒店、房地产等小范围场景的全景三维展示提供借鉴。     

基于数字微镜的旋转体扫描3维显示系统

体3维显示是真3维显示领域最前沿的一种解决方案,其优点是可以直接、多人、多角度同时观察。为了在真实空间显示复杂3维图像,研究并实现了一种基于数字微镜的被动式旋转体扫描3维显示系统,即首先利用人眼的视觉暂留特性,通过数字微镜将序列2维图像切片快速投影至旋转屏来完成真实3维图像的显示;同时分析了系统的组成和原理,并详细讨论了成像空间、体素激活装置和体3维引擎的硬件设计思想与软件核心算法。三棱锥和飞机实体模型的3维显示实验结果表明,用户可在360°范围内观察到轮廓清晰、立体感强的物体3维图像,说明该系统具备在真实空间显示复杂3维立体图像的能力,也说明方案是有效且实用的。     

古建筑视觉三维重建系统设计与实现

基于视觉的三维重建一直是计算机视觉领域的研究难点和热点,古建筑文化遗产的三维数字化应用就是其中之一.针对此需求,设计并实现基于图像视觉的古建筑三维重建系统.该系统由单个摄像机采集古建筑的多视角图像序列,通过图像特征提取和匹配、摄像机位置姿态计算、优化算法、三维点云生成、纹理映射等一系列自动计算过程,得到古建筑的三维重建模型.通过实验验证,该系统能够自动计算获取视觉效果良好的古建筑三维重建模型,且其实现成本低、操作简单安全,无须人工干预,可以满足当前古建筑快速安全的三维数字化应用需求.     

基于D3D立体镜显示器图像生成算法研究

介绍了自动立体镜显示器的结构及其三维显示的原理,研究了基于D3D技术的自动立体镜三维图像生成原理,对点蒙板方法做了深入的研究,给出了九片图中观察角度、位置以及视点观察空间的变换运算公式,在分析其优缺点的基础上提出了改进算法。并对最终图像的合成算法作了改进。在自动立体镜显示器上分别用原始方法、点蒙板方法、以及改进的点蒙板方法对图像进行显示,并在显示前准备时间、一帧图像渲染时间、三维深度感觉三方面对三者进行了对比实验。     

嵌入式μCGUI图像自适应滑动窗口的实现

目前μCGUI的高清显示存在较多问题,根源在于嵌入式操作系统资源有限,从而导致μCGUI无法很好地显示滑动显示图形。针对此问题,提出一种基于图像自适应滑动窗口算法与内存管理的μCGUI高清显示技术,大大降低了系统级的处理时间与内存消耗;并付诸实现在工程项目中,提升了基于嵌入式μCLinux操作系统的图形滑动显示的效果。     

手持式三维声纳实时成像系统设计

针对传统三维声纳成像系统集成度低、实时性差的问题,提出一种手持式三维声纳实时成像系统设计方案。利用DM8127实现声纳图像的处理,通过坐标转换数组预生成、自适应阈值的网格构建等方法实现实时高精度的三维建模,同时采用多线程并行处理架构改善网格构建时的系统处理性能,降低处理时延,并利用Open GL ES2.0实现三维声纳图像的渲染和显示。实验结果证明,该系统能清晰显示水下探测目标,且实时成像速率高达25帧/s,满足系统实时性要求。     

车载PEPS低频自动标定系统

针对车载PEPS系统的低频发射场实测问题,设计了一套汽车电子设备电磁场自动标定系统.该系统控制三维机械臂实现自动行走将场强测试仪送到指定测试点并记录场强值,然后通过可视化技术实现了电磁场强的三维分布显示.设计的标定系统实现了对PEPS系统低频发射场高效而精确的标定,为深入研究PEPS系统电磁场分布特征奠定了基础.     

Invited Paper: Depth reproducibility of multiview depth‐fused 3‐D display

Abstract— A multi‐view depth‐fused 3‐D (DFD) display that provides smooth motion parallax for wide viewing angles is proposed. A conventional DFD display consists of a stack of two transparent emitting screens. It can produce motion parallax for small changes in observation angle, but its viewing zone is rather narrow due to the split images it provides in inclined views. On the other hand, even though multi‐view 3‐D displays have a wide viewing angle, motion parallax in them is discrete, depending on the number of views they show. By applying a stacked structure to multi‐view 3‐D displays, a wide‐viewing‐angle 3‐D display with smooth motion parallax was fabricated. Experimental results confirmed the viewing‐zone connection of DFD displays while the calculated results show the feasibility of stacked multi‐view displays.     

基于多投影的多视点自动立体三维显示系统

多视点自动立体显示有望成为今后主流的三维显示技术,它是一种无需借助任何辅助观察设备的多视点、多观察区、高分辨率、显示效果逼真的三维显示方式。阐述了基于多投影的多视点自动立体显示系统的设计原理,详细地描述了系统的软硬件构架,建立了基于多投影仪和水平光学各向异性显示结构的自动立体显示样机,开发了投影仪阵列自动校准系统,提高了投影仪的校准精度,避免了因投影仪数目多而导致的繁琐的校准过程。实验结果能够给观众带来逼真的三维视觉体验。     

Influence of depth and 3D crosstalk on blur in multi‐view 3D displays

Current 3D crosstalk equation was defined from the characteristics of 3D display using glasses. This equation is not suitable for multi‐view 3D display with larger view number as it gives the inappropriately large value. In 3D display using eyeglass, double images occur at large depth. But, in multi‐view 3D display with larger view number, blur occurs to larger width for the larger depth. Hence, blur phenomenon of multi‐view 3D display was investigated to understand the unique characteristics of multi‐view 3D display. For this purpose, ray tracing S/W was used to simulate 3D display image seen at the designed viewing distance, to calculate the relative luminance distribution, and to quantify the relation between blur and depth. Calculated results showed that incomplete image separation caused the overlap of multiple view images and the blur. Blur edge width (BEW) was proportional to the horizontal disparity and related to the depth. BEWR = (BEW) / (binocular disparity) was newly defined, and its usefulness for 3D characterization was investigated. BEW and BEWR might be useful as new measuring items to characterize multi‐view 3D display regarding 3D crosstalk.     

Practical implementation of a depth‐feeling‐enhanced two‐plane electro‐floating display system using three‐dimensional integral images

Abstract— Although there are numerous types of floating‐image display systems which can project three‐dimensional (3‐D) images into real space through a convex lens or a concave mirror, most of them provide only one image plane in space to the observer; therefore, they lack an in‐depth feeling. In order to enhance a real 3‐D feeling of floating images, a multi‐plane floating display is required. In this paper, a novel two‐plane electro‐floating display system using 3‐D integral images is proposed. One plane for the object image is provided by an electro‐floating display system, and the other plane for the background image is provided with the 3‐D integral imaging system. Consequently, the proposed two‐plane electro‐floating display system, having a 3‐D background, can provide floated images in front of background integral images resulting in a different perspective to the observer. To show the usefulness of the proposed system, experiments were carried out and their results are presented. In addition, the prototype was practically implemented and successfully tested.     

矿物晶体标本的三维可视化展示设计与实现

矿物晶体标本具有很高的科研和观赏价值,但使用传统拍照技术在观赏和研究矿物晶体时,仍存在时间和空间上的限制。三维可视化展示技术可以使用户通过对鼠标或触摸的控制,实现上下、左右多个维度的移动和放缩,实现 360 度任意视角观察标本,极大地满足矿物晶体参观者和研究者的需求。本文介绍了目前三维可视化展示的技术方法,包括 WebGL 技术、全景技术、3D 打印技术和三维云展示技术。并且以国家岩矿化石资源共享平台矿物晶体标本的三维展示技术——环物摄影技术为例,阐述了三维展示技术在矿物晶体标本展示中的应用,详细介绍了环物摄影技术的设计过程与实现方式。     

Method and Realization of Calculating BRDF System with Multi\|angle Images

BRDF(Binomial Distribution Function) can describe the reflection characteristics of the surface,one of the cutting\|edge hot issues in multi-angle research in quantitative remote sensing.In order to study the reflection of the objects at different angles of observation,the measurement data are obtained by using the aerial photogrammetry method.Based on the collinear equation and semi-empirical core-driven model theory,the measurement of BRDF system with Multi\|angle Image is realized by using C# object\|oriented programming,including SFM(Struct From Motion) algorithm to solve camera parameters and establishment of the image pyramid to read image data efficiently and accurately.The system polarized graph were generated by selecting soil and maize ground point in the multispectral image data obtained by using the Micro\|MCA six-channel multi\|spectral camera,compared and analyzed with the image of the BRDF simulation software(DART).The results show that the system is computationally efficient and robust.The reflectivity of any observation angle extrapolated from the core driving model is in accordance with the characteristics of BRDF.The “hot spot” appears at the same side observation of the sun,which is consistent with the change trend of reflectivity of DART simulation image.The multi\|scale module designed in the system can meet the calculation requirements under different spatial resolution of image.The realization of the system provides a reference for the follow\|up study of multi\|angle measurement of BRDF in quantitative remote sensing.     

Distributed real-time rendering for ultrahigh-resolution multiscreen 3D display

Large-scale autostereoscopic three-dimensional (3D) displays can give audiences a truly immersive feeling with strong visual impact. However, the traditional autostereoscopic 3D display systems are limited by the display hardware, making it difficult to directly achieve large-scale 3D displays with high resolution. Multiscreen splicing with laser backlights can be used for large-scale and ultrahigh-resolution 3D display, but it normally results in subscreen image asynchronization, view zone error, or obvious edge overlapping. To solve the problems mentioned above, a distributed real-time rendering system for ultrahigh-resolution multiscreen 3D display is proposed. Fifteen 3D LCD display devices are driven through a host, cooperating with laser backlights, a lenticular lens array (LLA), and a directional diffuser to display high resolution, high frame rate, large size, and wide-viewing angle 3D images. The resolution of the whole display system can reach 23,040 × 21,600. The rendering system provides a large-scale and real-time 3D scene image with an ultrahigh-definition resolution at a speed of 40 frames per second and high quality.     

Circular camera array system for 3‐D displays based on the reconstruction of parallax rays

Abstract— A circular camera system employing an image‐based rendering technique that captures light‐ray data needed for reconstructing three‐dimensional (3‐D) images by using reconstruction of parallax rays from multiple images captured from multiple viewpoints around a real object in order to display a 3‐D image of a real object that can be observed from multiple surrounding viewing points on a 3‐D display is proposed. An interpolation algorithm that is effective in reducing the number of component cameras in the system is also proposed. The interpolation and experimental results which were performed on our previously proposed 3‐D display system based on the reconstruction of parallax rays will be described. When the radius of the proposed circular camera array was 1100 mm, the central angle of the camera array was 40°, and the radius of a real 3‐D object was between 60 and 100 mm, the proposed camera system, consisting of 14 cameras, could obtain sufficient 3‐D light‐ray data to reconstruct 3‐D images on the 3‐D display.     

基于单投影仪与柱面反射镜的沉浸感显示系统

在分析传统沉浸感显示系统优缺点的基础上,设计并实现了一种新型的沉浸感显示系统。该系统使用柱面反射镜对单投影仪投射光线进行反射,并在弧形背投幕上成像,能够获得无缝画面,实现了广角度虚拟场景的连贯显示。通过合理设计柱面反射镜形状,实现了投影画面在水平方向上的均匀放大。通过对图像进行预变形可以基本消除因投影幕曲率而产生的投影画面形变,同时采用背投技术使参与者在虚拟场景中活动更加自如。该系统克服了传统的多投影仪或多显示器沉浸感显示系统中存在的图像拼接问题,且易于构建,经实验验证能够获得良好的沉浸感显示效果。     

Influence of 3‐D cross‐talk on qualified viewing spaces in two‐ and multi‐view autostereoscopic displays

Abstract— To estimate the qualified viewing spaces for two‐ and multi‐view autostereoscopic displays, the relationship between image quality (image comfort, annoying ghost image, depth perception) and various pairings between 3‐D cross‐talk in the left and right views are studied subjectively using a two‐view autostereoscopic display and test charts for the left and right views with ghost images due to artificial 3‐D cross‐talk. The artificial 3‐D cross‐talk was tuned to simulate the view in the intermediate zone of the viewing spaces. It was shown that the stereoscopic images on a two‐view autostereoscopic display cause discomfort when they are observed by the eye in the intermediate zone between the viewing spaces. This is because the ghost image due to large 3‐D cross‐talk in the intermediate zone elicits different depth perception from the depth induced by the original images for the left and right views, so the observer's depth perception is confused. Image comfort is also shown to be better for multi‐views, especially the width of the viewing space, which is narrower than the interpupillary distance, where the parallax of the cross‐talking image is small.