Improvement of Camera Calibration Accuracy Based on Periodic Arrangement Characteristics of Calibration Target Pattern

Conventional camera calibration that employs calibration targets is a commonly used method to acquire a camera's intrinsic and/or extrinsic parameters. The calibration targets are usually designed as periodic arrays of simple high-contrast patterns that provide highly accurate world coordinate system points and the corresponding image pixel coordinate system points. The existing pixel coordinate extraction algorithms can reach a sub-pixel level; however, they treat each single pattern in one image as an independent individual, which makes it difficult to further improve extraction accuracy. In this paper, a novel method is proposed by utilizing the periodic arrangement characteristics of the calibration target pattern as a global constraint to improve the calibration accuracy. Based on a camera's pinhole model, the intersection point of two fitted curves is used as an optimized pixel point to replace the initial one. Following the pixel coordinate optimization procedures, experiments were performed using real data from a 3D laser line scanner and a dynamic precision calibration target. Our results show that the relative errors of camera homography matrix elements obtained by the proposed optimization method were reduced compared with the commonly used method. The average coordinate measurement accuracy can be improved by nearly 0.05 mm. It is shown that the proposed optimization method can enhance the camera calibration accuracy,especially when the extracted pixels are of poorer precision.     

室内图书运载车的计算机视觉定位定向方法

提出一种可用于图书馆内图书运载车的车辆室内自动定位定向方法.借鉴二维条形码思想,设计了一种简单方便、易于识别、内含绝对位置坐标、具有一定纠错能力的路标,可将路标放置在图书馆天花板上,由安装于图书运载车上,且光轴和天花板垂直的摄像机进行拍摄,通过图像分割,连通域寻找,轮廓特征曲线匹配以及路标特征识别步骤定位路标,解析路标中所包含的坐标信息以及偏转角度,进而计算出车辆当前所处的绝对位置和航向角.通过实验,验证了所提出方法的有效性,能够满足图书馆内图书运载车室内定位的要求.     

复杂曲面轮廓误差高精度在线计算研究

高精度在线计算轮廓误差可有效降低数控加工中的控制轮廓误差。因此,提出一种基于三点弧线理论的方法,在补偿轮廓误差之前对轮廓误差进行精确计算。首先根据插值参考位点与加工过程中的实际位置,计算距离实际位置最近的参考位点;然后根据最近参考位置及其相邻两点形成的近似外接圆弧,计算实际位置到该圆弧的距离,即求得刀尖位点轮廓误差,采用同样方法计算刀具方向轮廓误差。实验结果表明,该算法比原有算法刀尖位点轮廓误差均值减小了 0.056um,刀具方向均值比原有算法减少 7.166um。基于三点弧线理论的算法仅需考虑实际位置与距离实际位置最近的参考位点坐标,可精确计算较大的曲率处,且无需考虑传统计算方法中刀具干涉问题,简化计算过程的同时还可保证计算精度。     

数码相机对三维空间尺度测距研究

提出了一种由单幅图像得到目标物体深度信息的方法。该方法通过计算机图像处理技术获取相关的图像信息,再结合光学成像原理将所用数码相机的实际物理焦距和图像中目标物体的大小均效到135相机的标准上,由计算机编程计算得出目标与测量系统之间的相对位置关系。介绍了测量原理,推导出了参数计算公式。实验表明,此系统测量精度满足工程测量要求。     

基于单目视觉的单图标定算法

为了满足现代工业生产中精度要求高、操作快捷的要求,提出基于单目视觉的单图标定算法。通过采集一张棋盘格标定板图像,精确提取棋盘格角点坐标,使用仿射变换得到像素坐标系和世界坐标系的仿射变换矩阵,分析Halcon仿射变换矩阵和张氏标定法相机模型的对应关系,得到详细的相机内参,再对图像进行畸变矫正,得到像素当量。多次Halcon实验结果表明,提出的算法与其他方法相比,操作更加快捷,标定结果标准差可控制在0.02 mm 以内,能够满足现代工业生产需要。     

疏港航道大桥基桩导线测量放样的控制

测量放样的控制直接影响工程施工放样的准确性。结合大丰港疏港航道大桥基桩测量放样工作,采用导线测量确定各加密点的坐标和高程的具体测设过程。导线测量结果构成了本项目施工测量控制网,确保了基桩施工放样的精确性。     

Environment map building and localization for robot navigation based on image sequences

SLAM is one of the most important components in robot navigation. A SLAM algorithm based on image sequences captured by a single digital camera is proposed in this paper. By this algorithm, SIFT feature points are selected and matched between image pairs sequentially. After three images have been captured, the environment＇s 3D map and the camera＇s positions are initialized based on matched feature points and intrinsic parameters of the camera. A robust method is applied to estimate the position and orientation of the camera in the forthcoming images. Finally, a robust adaptive bundle adjustment algorithm is adopted to optimize the environment＇s 3D map and the camera＇s positions simultaneously. Results of quantitative and qualitative experiments show that our algorithm can reconstruct the environment and localize the camera accurately and efficiently.     

盾构位姿测量方法的研究

在地下暗挖施工过程中,盾构机的掘进路线必须在设计误差要求范围内,因此实时监控盾构的运动状态至关重要。作为盾构机眼睛的导向系统要求高精度、实时快速地测量出盾构的位置和姿态。基于嵌入式技术设计的激光位移检测靶主要由CCD照相机、倾角传感器和棱镜等光学元件组成,结合全站仪和后视棱镜,通过CCD采集全站仪发射的激光光斑,对光斑图像进行去噪、形态学等处理,准确计算中心点坐标,通过坐标系的转换,最终实现对盾构的滚动角、方位角和俯仰角的测量。     

导线测量在道路施工中的应用

导线测量是一项精密而细致的工作,直接影响到路线的定线和高程系统的控制。结合连云港市新港大道XL-5标段,采用全站仪导线测量确定各加密点的坐标和高程的具体测设。导线测量结果构成了本项目施工测量控制网,确保了工程施工放线的精确性。     

Vision based terrain reconstruction for planet rover using a special binocular bundle adjustment

This paper presents a pure vision based technique for 3D reconstruction of planet terrain. The reconstruction accuracy depends ultimately on an optimization technique known as ＇bundle adjustment＇. In vision techniques, the translation is only known up to a scale factor, and a single scale factor is assumed for the whole sequence of images if only one camera is used. If an extra camera is available, stereo vision based reconstruction can be obtained by binocular views. If the baseline of the stereo setup is known, the scale factor problem is solved. We found that direct application of classical bundle adjustment on the constraints inherent between the binocular views has not been tested. Our method incorporated this constraint into the conventional bundle adjustment method. This special binocular bundle adjustment has been performed on image sequences similar to planet terrain circumstances. Experimental results show that our special method enhances not only the localization accuracy, but also the terrain mapping quality.     

泊车机器人障碍物视觉识别系统研究

针对智能停车库中的泊车机器人视觉系统研究需求,提出一种基于双目视觉的泊车机器人障碍物识别系统。通过双目摄像头进行图像采集,利用张正友棋盘标定法进行双目相机标定;采用Bouguet进行立体校正,将高斯滤波与拉普拉斯算子相结合进行图像预处理;采用YOLO卷积神经网络对目标障碍物进行快速识别;利用区域匹配算法进行立体匹配并生成目标障碍物视差图;通过成像点和目标障碍物的立体几何关系计算得到目标障碍物的深度信息。实验结果表明,该系统具有良好的实时性和较高精度,障碍物识别时间平均为0.0901s,在2 600mm具有最佳测距精度,可为泊车机器人自动泊车提供保障。     

CASIO fx-4500P计算器在建筑施工变形监测测边交会法中的应用研究

测边交会法可用于建筑施工变形监测中,传统的方法是钢尺丈量边长交会法,随着电子技术的发展,电子测距仪的应用,特别是全站仪的逐步推广与应用,测边交会法用于建筑施工变形监测又有广泛应用的趋势。但交会点的坐标及误差计算比较复杂,经作者进行测边交会的误差分析的计算结果公式,应用可编程序型计算器fx-4500P及开发的程序,可快速、准确计算测边交会点的坐标和点位精度。     

井下无定向导线测量

在巷道延伸测量中，经常会遇到测量控制点被破坏，剩下的点又彼此不通视的情况，为了节省作业时间，不影响生产，应用井下无定向导线测量，可以恢复该段导线；该法是把两已知控制点看着是定向时两垂球线连接点，在两控制点间按所设计的精度进行相应等级的导线联测，并用两井几何定向的井下连接导线解算方法进行计算，使原来按一般导线测量方法不能解决的点位坐标问题得以解决。     

基于视觉引导的机器人抓取分类系统设计

为了解决流水线的快速抓取及分类问题,采用六自由度串联工业机器人和SCARA 四轴机器人结合,配合3D 和2D 相机,搭建了一套智能分类抓取系统。基于Halcon 视觉处理平台,在Qt 软件框架下进行二次开发,构建了机器人视觉引导的自动抓取和分类软件框架。通过相机采集图像,并在上位机中经图像预处理、位姿估计、模板匹配等步骤后,得到三维位姿或中心坐标,并发送给机器人实行智能抓取,从而实现对多种堆叠物块的识别和抓取。实验结果表明,通过视觉定位得到的物块中心位置和机械手实际抓取测量之间的误差为0.05~1.22 mm,摆放角度在5°以内时,分拣效率比人工分拣提高了62%左右,可以更好地满足实际企业生产的需求。     

激光测量技术在热态成品车轮尺寸测量中的应用

激光三角法测量原理是目前光学测量应用最广泛的方法之一。其原理是光源发生器发出结构光光源（一般为激光点光源）到被测工件的表面,通过光学系统成像到线阵CCD或面阵CCD上的住置,由几何三角关系,从像点在CCD上的位置即可计算得到工件的高度尺寸,再通过测量系统的测量运动（即扫描运动）得到工件的全部外形尺寸。着重介绍热态成品车轮激光检测装置的结构配置及其完成的主要功能。     

Flexible Calibration Method for 3D Laser Scanner System简

In this paper, a flexible high-precision calibration method suitable for industrial field was proposed. The complexity of the coordinate transformation was simplified by choosing the camera coordinate system as the unified reference coordinate system. A flexible planar calibration pattern was introduced to the calibration process, which can be arbitrarily placed and from which the known feature points can be extracted to construct other unknown feature points. With the known intrinsic parameters, the laser projector plane equation was fitted by the multi-noncollinear points, which were acquired through the principle of triangulation and the projective invariance of cross ratio. With this method, the strict alignment and multiple times of coordinate transformation can be avoided. Experimental results showed that the arithmetic mean of the root mean square （RMS） error of distance was 0.000 7 mm.     

双目立体视觉定位的三维坐标提取与分析

双目定位跟踪系统涉及到的关键技术有摄像机标定、立体匹配和三维重建等.在三维重建阶段,利用匹配的结果采用最小二乘法计算定位标靶上标记的三维坐标.实验结果表明,在现有条件下,双目定位跟踪系统的定位精度比较高,相对误差在1%以内;基于此定位跟踪系统,能够初步准确完成对经络的三维定位和实时显示,为临床诊断治疗和基础教学提供直观的分析依据.     

Precise navigation for a 4WS mobile robot

INTRODUCTION Precise navigation is being researched and ap- plied in various fields, especially for outdoor mobile robots (Wang et al., 1998; Lee, 1997; Iida et al., 2000), such as those used in cargo container trans- portation at seaport, in precision agriculture, military vehicles, etc. As we know, applied sensors and rele- vant data processing basically determine the accuracy of navigation. In our research, GPS and encoders are used as navigation sensors. As an absolute positioning s…     

基于机器视觉的茄科幼苗切削特征识别

为了提升蔬菜自动嫁接机嫁接自动化程度,在嫁接过程中实现机器替代人眼进行操作,基于机器视觉分析茄科幼苗轮廓链特征参数、幼苗轮廓链链角变化、跨距、水平截线宽度、幼苗估计苗茎及幼苗中心线斜率参数,确定幼苗子叶、真叶和根部位置,根据幼苗轮廓特征结合曲线拟合确定切削点位置和切削角度,使用100幅蕃茄幼苗图像进行实验。实验表明,基于机器视觉的幼苗切削参数特征识别及定位准确率较高,幼苗轮廓链提取准确为97%,切削点位置与切削角度确定的准确率分别为93%和90%。因此基于机器视觉可以快速、准确地确定切削点位置及切削角度,提高嫁接机嫁接过程中嫁接操作准确率及幼苗成活率。