RGB-D SLAM在室内高精度三维测图中的应用

高精度三维测图是室内三维制图的重要支撑,基于三维激光雷达扫描技术的三维测图成本高,需要提前布置标靶,在室内复杂环境中易导致数据不完整;基于图像序列的三维重建建模时间长,易受多种因素影响。针对以上问题,本文将RGB-D SLAM技术应用于室内高精度三维测图中。通过将深度相机与SLAM技术相结合,计算相机位姿并恢复三维空间信息,获取室内三维点云模型,并以目标物实际量测为基准评价密集点云精度。试验结果表明,该方法可快速获取精度较高的三维点云模型,成本低且效率高,能够较好地满足应用需求。     

增强现实地图研究与应用

随着数字地图技术的不断革新,产生了将虚拟信息与真实世界融合的新型地图技术,即增强现实(augmented reality,AR)地图。从可视化、定位、测绘、分析几个方面阐述了AR地图的关键技术,介绍了AR地图的空间映射原理和可视化效果,提出了结合视觉定位和惯导定位的cm级室内定位方法,实现了基于AR地图的AR导航、AR测量、AR目标采集、AR地理围栏等关键技术并将其应用于AR智慧园区与AR室内场馆。     

基于移动LiDAR点云数据的室内全要素三维重建

精确的室内三维模型在古遗产保护、公共场所的精细化管理和室内导航等领域有着重要作用,而激光雷达正被逐步应用到室内三维重建中。文中采用移动激光雷达对室内所有附属设施进行高精度激光扫描,得到融合CCD影像色彩信息的彩色点云数据。给出室内全要素三维重建的技术流程,并利用Autodesk CAD等三维建模平台对点云数据进行有效处理,最终得到高精度且带有大地坐标的室内三维模型。实验结果表明,采用移动激光雷达能快速获取高精度的室内三维点云数据。通过文中提出的流程,可快速得到真三维、真尺寸、真纹理的室内全要素模型。     

融合RGB-D深度图像和LiDAR点云的石油管线BIM建模

建筑信息模型(BIM)已经被广泛用于提高复杂管线系统的管理效率,但已有石油管线的BIM建模,往往依靠设计图纸或现场测绘手工完成,耗时费力。为此本文提出了一种融合RGB-D深度图像和LiDAR点云数据的石油管线BIM重建方法。首先利用RGB-D图像提供的丰富语义信息和LiDAR点云精确几何信息,对深度相机采集的RGB图像进行分割,生成三维语义地图;然后通过点云粗匹配和精确匹配实现数据融合;最后给出了不同结构管线构件的BIM模型制作方法。试验结果表明,与以往的管线BIM重建方法相比,该方法更准确、高效,有助于石油企业对含有复杂管线的计转站等实施信息化管理。     

低成本大视场深度相机阵列系统

在测距传感器不断轻量化、小型化以及室内外地图一体化导航应用的驱动下,三维（3D）室内移动测量成为当今研究和应用的热点,在室内建模、室内定位等新兴领域中的应用越来越广泛。3D室内移动测量系统通常配备激光扫描仪、全景相机、惯性测量单元（inertial measurement unit,IMU）系统和里程计等传感器,虽能实现3D室内点云数据的采集,但其距离传感器-激光扫描仪价格昂贵且便携性较差。彩色深度（RGB depth,RGB-D）相机为低成本3D室内移动测量系统构建提供了新的距离成像传感器选择,但主流型号RGB-D相机视场角小,继而导致数据采集效率远低于传统激光扫描仪,难以做到点云数据的完整覆盖与稳健采集,且易造成同时定位与制图（simultaneous localization and mapping,SLAM）过程中跟踪失败。针对以上问题,构建了一种低成本室内3D移动测量系统采集设备,通过组合多台消费级RGB-D相机构成大视场RGB-D相机阵列,提出了一种阵列RGB-D相机内外参数标定方法,并通过实验检验了设计系统采集的点云数据的精度。     

多镜头组合式相机的全景SLAM

同步定位与地图构建(simultaneous localization and mapping,SLAM)是摄影测量、计算机视觉和机器人学的研究热点,并广泛应用于移动测图系统、机器人、无人驾驶车等。本文提出一种针对多镜头组合式全景相机的基于特征的SLAM解决方案。首先,本文建立了鱼眼相机的高精度检校模型,以保证鱼眼相机与全景相机之间的高精度坐标转换;然后,将多镜头组合式全景成像模型嵌入SLAM的初始化、局部地图生成、关键帧选取、图优化等各个流程中。此外,考虑全景相机变形大、基线长的不利因素,本文在特征匹配、平差、特征点跟踪等SLAM的各个步骤都进行了针对性改进。本文在两套车载全景数据集共8000余张全景影像上进行试验。结果表明,本文所提出的全景SLAM很好地实现了全景相机的自动定位与地图构建功能,并达到了接近GPS参考的极高定位精度而无须借助GPS/IMU组合导航系统。相对于主流的基于平面相机的各类SLAM系统,如Mono-SLAM、Stereo-SLAM以及RGB-DSLAM,本文提出的全景SLAM可作为良好的补充,并为GPS信号失锁时的传感器定位提供廉价的自动解决方案。     

预建高精度地图的封闭区域UGV自动驾驶导航定位

针对封闭区域UGV自动驾驶应用,本文提出了一种基于平面高精度地图的导航定位方法。该方法利用三维激光扫描数据采用预定义的栅格地图概率值建立多分辨率地图,在保证定位精度的同时提高定位效率,采用极大似然估计获取载体位姿初值并将IMU数据用于计算高斯-牛顿法搜索初值。试验结果表明,基于激光扫描的二维地图构建与匹配定位方法能有效解决帧与帧匹配误差快速累积问题,可以高效地使用已有地图进行连续高精度的载体定位。     

激光SLAM移动机器人室内定位研究

针对目前室内移动导航定位精度低和累积误差大的问题,提出了一种激光雷达（LiDAR）和惯性测量单元（IMU）相融合的导航定位系统。首先,该方法是从LiDAR扫描测量中提取环境特征和构建地图,然后,由IMU采集的姿态信息通过卡尔曼滤波,补偿由于LiDAR扫描引起的位置和姿态输出的误差,以提高机器人移动的定位精度。试验结果表明,该方法可以提高室内移动机器人定位和构建地图的精度和稳健性。     

Augmented reality and photogrammetry: A synergy to visualize physical and virtual city environments

Close-range photogrammetry is based on the acquisition of imagery to make accurate measurements and, eventually, three-dimensional (3D) photo-realistic models. These models are a photogrammetric product per se. They are usually integrated into virtual reality scenarios where additional data such as sound, text or video can be introduced, leading to multimedia virtual environments. These environments allow users both to navigate and interact on different platforms such as desktop PCs, laptops and small hand-held devices (mobile phones or PDAs). In very recent years, a new technology derived from virtual reality has emerged: Augmented Reality (AR), which is based on mixing real and virtual environments to boost human interactions and real-life navigations. The synergy of AR and photogrammetry opens up new possibilities in the field of 3D data visualization, navigation and interaction far beyond the traditional static navigation and interaction in front of a computer screen.In this paper we introduce a low-cost outdoor mobile AR application to integrate buildings of different urban spaces. High-accuracy 3D photo-models derived from close-range photogrammetry are integrated in real (physical) urban worlds. The augmented environment that is presented herein requires for visualization a see-through video head mounted display (HMD), whereas user’s movement navigation is achieved in the real world with the help of an inertial navigation sensor. After introducing the basics of AR technology, the paper will deal with real-time orientation and tracking in combined physical and virtual city environments, merging close-range photogrammetry and AR. There are, however, some software and complex issues, which are discussed in the paper.     

基于图优化的紧耦合双目视觉/惯性/激光雷达SLAM方法

基于单一传感器的同时定位与地图构建技术已经逐渐不能满足移动机器人、无人机及自动驾驶车辆等智能移动载体日益复杂的应用场景。为了进一步提升移动载体在复杂环境下的定位与建图性能,基于多传感器融合的SLAM技术成为目前研究的热点内容。本文提出了一种基于图优化的紧耦合双目视觉/惯性/激光雷达SLAM方法(S-VIL SLAM),该方法在视觉惯性系统中引入激光雷达原始观测,基于滑动窗口实现了IMU量测、视觉特征及激光点云特征的多源数据联合非线性优化。利用视觉与激光雷达的互补特性设计了视觉增强的激光雷达闭环优化算法,进一步提升了多源融合SLAM系统的全局定位与建图精度。为了验证本文算法的性能,利用自主搭建的集成多传感器的硬件采集平台在室外场景下进行了车载试验。试验结果表明,本文提出的紧耦合双目视觉/惯性/激光雷达里程计相比于紧耦合双目视觉惯性里程计和激光雷达里程计定位定姿性能显著提升,视觉增强的激光雷达闭环优化算法能够在大尺度场景下有效探测出轨迹中的闭环信息,并实现高精度的全局位姿图优化,经过闭环优化的点云地图具有良好的分辨率和全局一致性。     

无人机倾斜摄影与LiDAR融合监测技术

本文以成都市环城生态区生态修复项目为依托,利用飞马D200无人机实现了倾斜摄影和LiDAR技术的融合监测,通过三维模型制作、大比例尺地形图生产、点云处理、方格网计算、精度评定等步骤,验证了倾斜摄影和机载LiDAR协同监测方式可以满足1∶500地形图和方格网测量的精度,对后期无人机测绘实际生产具有指导意义。     

基于几何约束和目标检测的室内动态SLAM

针对即时定位与地图构建(SLAM)在室内动态环境下定位精度低和地图效果差的问题,提出一种基于几何约束和目标检测的室内动态SLAM方法.使用目标检测网络获取语义信息,提出运动物体漏检的方法;根据先验知识,提出准确识别动态区域的信息判定方法;结合几何约束和深度学习方法剔除动态点,利用静态点估计相机位姿;根据存储信息构建可闭环的静态地图.在TUM数据集上进行实验,定位精度比ORB-SLAM2提高97.5%,相较于其他动态SLAM可取得更好的性能.在室内真实环境进行实验,构建的静态地图更准确,有效提高了室内动态SLAM的定位精度和地图效果.     

Colour Hue and Texture Evaluation for 3D Symbolization of Indoor Environments Using RGB-D Data

In this work an evaluation procedure of the use of colour hue and texture for 3D symbolization of indoor environments using RGB-D data is presented. The main characteristics of the proposed method involve RGB-D cameras to collect colour and depth data and the evaluation of the visual variables colour hue and texture combined with visual aspects such as lighting and camera configuration for the symbolization of 3D indoor environments. High quality coloured point clouds of indoor environments are obtained with the Kinect sensor and then registered and processed in order to create 3D models. These sets of point clouds can be detected, recognized and classified under one of the three following main categories: ceiling, wall or floor. After classification is done, the resulting point clouds are properly treated as 3D triangular meshes in the MeshLab software. These meshes are then filtered to form complete surfaces which are then refined before being considered for visual variable application. We evaluated the proposed method on several indoor environments to show that it efficiently combines the colour and depth data obtained from RGB-D cameras with different parameters defined for the visual variables colour hue and texture for 3D symbolization of indoor environments.     

机载LiDAR技术支持下的荒漠地区1:10 000地形图测绘与研究——以内蒙古阿拉善地区为例

尝试应用机载LiDAR技术测绘1：10 000比例尺地形图3D（DLG、DEM、DOM）产品,给出了机载LiDAR测绘3D产品的技术流程,并选择荒漠地区作为试验区,验证了此种技术方法在荒漠地区测绘3D产品的可行性,分析了成果精度。试验证明,该方法可以满足荒漠区域的1：10 000比例尺3D基础数据生产要求,且具有外业工作量小、自动化程度高、成图快、高程精度高、受外界环境影响小等优点,同时也总结了该方法中有待完善之处。该方法为荒漠地区3D基础测绘数据获取提供了有益借鉴。     

A literature synthesis of LiDAR applications in transportation: feature extraction and geometric assessments of highways

Interest in using Light Detection and Ranging (LiDAR) technology in Transportation Engineering has grown over the past decade. The high accuracy of LiDAR datasets and the efficiency by which they can be collected has led many transportation agencies to consider mobile LiDAR as an alternative to conventional tools when surveying roadway infrastructure. Nonetheless, extracting semantic information from LiDAR datasets can be extremely challenging. Although extracting roadway features from LiDAR has been considered in previous research, the extraction of some features has received more attention than others. In fact, for some roadway design elements, attempts to extract those elements from LiDAR have been extremely scarce. To document the research that has been done in this area, this paper conducts a thorough review of existing studies while also highlighting areas where more research is required. Unlike previous research, this paper includes a thorough review of the previous attempts at data extraction from LiDAR while summarizing the detailed steps of the extraction procedure proposed in each study. Moreover, the paper also identifies common tools and techniques used to extract information from LiDAR for transportation applications. The paper also highlights common limitations in existing algorithms that could be improved in future research. This paper represents a valuable resource for researchers and practitioners interested in knowing the current state of research on the applications of LiDAR in the field of Transportation Engineering while also understanding the opportunities and challenges that lie ahead.     

船载移动激光三维测量系统安置角偏差检校

船载移动激光三维测量系统是集激光扫描仪、全球定位系统和惯性导航测量单元等于一体的多传感器集成系统,具有效率高、精度高、三维测量等特点,可解决码头、河道、海岛礁测绘中传统方法难以测量的难题。船载移动激光三维测量系统的高精度时空同步是实现数据融合和高精度三维测量的保证,安置角偏差检校是船载移动激光三维测量的关键步骤。本文首先分析了船载移动激光三维测量系统中相关坐标系之间的转换关系,提出了一种以桥体为检校场的安置角偏差消除方法。该方法在时空配准的基础上,通过扫描测线内桥体结构的偏移量分别计算侧滚角、俯仰角及航向角的安置角偏差,最后通过控制点验证误差改正后的测量精度。试验结果证明,本文方法易于实施,且具有较高的准确性,能够有效保障船载移动激光测量数据的质量和精度。     

移动AR+VR支持下旅游GIS系统的设计与实现

随着云计算、物联网、移动通信、移动互联网等技术的发展,研发智能化、多维度、全方位的旅游管理信息系统,已成为旅游信息化的发展趋势。本文以青岛八大关为例,综合利用基于位置与多传感器的跟踪注册技术、瓦片地图技术、三维景观地图技术和移动增强现实系统虚实叠加技术,研究并实现了面向旅游业的移动化、信息化、智能化、个性化的智能旅游服务系统,为游客提供认知周围景观的全新视角,获得全新的交互体验。系统包括三维景观地图、虚拟漫游、路线导航、环境识别、历史建筑查询、热力图显示、好友足迹、个人轨迹、微信分享等功能模块,游客在享受旅途的同时可以获得更加智能便捷的服务。     

Indoor scene texturing based on single mobile phone images and 3D model fusion

Realistic texture mapping and coherent up-to-date rendering is one of the most important issues in indoor 3-D modelling. However, existing texturing approaches are usually performed manually during the modelling process, and cannot accommodate changes in indoor environments occurring after the model was created, resulting in out-dated and misleading texture rendering. In this study, a structured learning-based texture mapping method is proposed for automatic mapping a single still photo from a mobile phone onto an already-constructed indoor 3-D model. The up-to-date texture is captured using a smart phone, and the indoor structural layout is extracted by incorporating per-pixel segmentation in the FCN algorithm and the line constraints into a structured learning algorithm. This enables real-time texture mapping according to parts of the model, based on the structural layout. Furthermore, the rough camera pose is estimated by pedestrian dead reckoning (PDR) and map information to determine where to map the texture. The experimental results presented in this paper demonstrate that our approach can achieve accurate fusion of 3-D triangular meshes with 2-D single images, achieving low-cost and automatic indoor texture updating. Based on this fusion approach, users can have a better experience in virtual indoor3-D applications.     

车载MMS激光点云与序列全景影像自动配准方法

提出一种车载移动测量系统（MMS）激光点云与序列全景影像自动配准方法。首先采用层次化城市场景目标提取方法自激光点云提取天际线矢量,在全景影像中经虚拟成像与分割角点提取算法生成天际线矢量。然后,将提取结果作为几何配准基元,构建配准基元图,通过最小化配准基元图编辑距离进行匹配,组成共轭配准基元对,解算2D-3D粗配准模型,获得全景影像与LiDAR点云参考坐标系之间的初始转换关系。最后,为消除几何配准基元提取与匹配误差对配准结果的影响,自序列全景影像虚拟成像影像生成多视立体密集匹配点云,继而使用变种ICP算法优化其与激光点云数据间3D-3D配准参数,间接优化全景影像与激光点云间的配准参数,精化配准结果。试验结果表明,本文提出的自动配准方法可以实现车载MMS激光点云与序列全景影像的1.5像素级自动配准,配准成果可应用于真彩色点云生成等点云/影像数据融合应用。     

移动端AR技术在海洋地貌可视化中的应用

将移动端增强现实(AR)技术引入海洋地貌可视化中,从海洋洋体和海底地貌两个方面探讨了海洋地貌AR可视化应用系统的设计思路和实施方案。首先将某一海域的海底地貌进行标准化处理,利用Blender建立海洋三维地貌模型;再基于Unity设计实现面向移动端的海洋地貌AR可视化交互系统,突破了目标识别、跟踪注册、坐标转换等关键技术,实现了放大、缩小、旋转、视角变换以及海洋洋体与海底地貌模型属性修改等人机交互功能,将真实海洋世界与虚拟海洋世界相结合,取得了较好的应用效果。