基于视觉的多点触摸基本技术实现方法

高性能的图像处理技术是基于视觉的多点触摸技术的关键,对于特征点的检测和跟踪是完成多点触摸系统的基础.文中主要介绍了几种算法用来解决特征点的识别和跟踪的问题.首先,图像轮廓变换算法用来分析手指与屏幕的接触区域,以及在此基础上采用中心计算算法确定特征点的中心坐标;再次,最小距离优先(MDF)算法对在两个相邻图像中的相应的特征点进行识别和跟踪,为随后的事件检测和手势识别提供了前提条件.实验结果表明,算法的性能满足多点触摸系统的实时要求.     

增强现实中的摄像机径向畸变校正

增强现实系统中,有效地进行摄像机镜头畸变校正对提高虚拟环境的精确性具有重要意义.首先提出一种基于成像几何的畸变校正方法,采用带有一阶径向畸变的摄像机模型,对镜头径向畸变进行校正,再根据校正后的图像计算摄像机投影矩阵.实验表明,基于成像几何的畸变校正算法具有较高的畸变校正精度,640×480的图像中,最大畸变量达90多个像素.与利用理想针孔摄像机模型得到的增强现实环境相比,畸变校正后得到的叠加结果更为精确.     

基于等间距线的几何畸变校正

摄像机视觉测量系统中图像存在几何畸变,会降低尺寸测量精度,必须进行校正.目前基于单张图像的畸变校正方法主要有标定板特征点转移矩阵的非参数化方法和利用特征点共线性或无穷点约束条件的参数化方法,由于上述方法需要获取多个特征点的坐标,而获取孤立的坐标点容易造成误差,使得校正效果不好.提出采用等间距线方法,利用畸变模型建立间距...     

运动摄像机下多目标检测与跟踪

针对摄像机运动情况下多目标的检测与跟踪问题,提出一种将Global K均值与模板匹配相结合的方法.利用六参数仿射模型得到摄像机运动参数,对图像进行全局运动补偿,用GlobalK均值算法对前景点进行循环聚类,判断目标数目并进行跟踪,通过对目标区域进行模板匹配使跟踪结果更准确.实验结果表明,该方法能够在运动摄像机下稳定、实时地跟踪多个目标,对发生形变的目标基本也能稳定跟踪.     

基于深度图像的运动人手检测与指尖点跟踪算法

针对由传统摄像头获得的二维图像难以准确检测复杂环境下运动人手目标的问题,提出了一种利用Kinect摄像头对复杂背景下近景人手指尖空间点的检测和跟踪算法。首先利用基于视觉的多信息融合方法对人手区域进行粗检测,并利用深度信息对得到的模糊人手区域进行精确定位及分割,从而有效抑制了复杂背景对人手区域的干扰;根据分割得到的人手区域,利用人手边界的曲率关系实现了指尖空间点的快速检测;最后利用卡尔曼滤波跟踪检测到的运动指尖点。实验结果表明,与已有指尖点检测方法相比,该算法能够准确地跟踪多种姿态下手指指尖的空间位置,具有较低的误差率。     

基于直线投影特征的镜头畸变校正方法

针对摄像机镜头畸变校正方法的简便和快速问题,设计了基于直线投影特征的校正方法。介绍了镜头主要畸变产生原因和畸变模型;给出直线三点在理想投影下的关系,确定了适应度函数,利用遗传算法得到了畸变参数组最优解。基于matlab软件编写校正程序,并进行了实验验证。实验表明,利用畸变参数组的最优解能够实现图像畸变校正,效果较好。该标定方法只需场景内有直线存在即可实现对摄像机镜头畸变参数校正,方法所需实验条件简单,程序简便,便于现场快速校正。     

基于VRML的虚拟交互系统触摸感知控制仿真研究

针对虚拟交互系统触摸感知控制过程中图像存在噪声干扰,导致触摸感知效果下降的问题,提出了基于VRML的虚拟交互系统触摸感知控制方法。将初始的三维立体空间图像转换为二维平面空间图像,将不间断的图像加以变换,得到离散型的数据,通过图像中不同像素点的平均值来取代该点的灰度值,得到无噪声数字图像。针对图像中任意像素点,对其边缘点进行检测,完成边缘图像细节信息识别获取。在上述基础上,对目标图像进行轮廓提取,采用改进后的归一化傅里叶描述子对经过去噪处理的图像进行三维识别,利用VRML重建识别结果,实现虚拟交互系统触摸感知。通过具体的仿真数据充分验证了所提方法的综合有效性。     

大视场景物非线性畸变校正的仿真

图像在应用时往往需要物体在图像中准确的位置,但由于物体在从世界坐标系到摄像机坐标系变换时存在平移和旋转,所以拍摄的图像会存在非线性几何畸变.为此针对广角镜头非线性畸变,在分析多项式变换的基础上,提出多项式分区域校正方法.首先检测标准网格和畸变网格的坐标,计算出镜头的畸变系数;然后拟合多项式进行坐标修正,重建图像;最后对重建的图像进行分块,计算每个小块的畸变系数并再次重建图像.实验结果表明,提高了图像校正的精度和智能化,从而达到高质量校正的目的.     

一种基于斜率的摄像机畸变校正方法

普通 CCD摄像机在成像时都存在畸变成像误差 ,在机器人视觉检测及自动装配中 ,有效地进行误差校正对准确确定物体的位置具有重要的意义 .本文采用带有一阶径向畸变的小孔摄像机模型 ,提出一种基于线段斜率的方法 ,对摄像机镜头的径向畸变进行校正 ,不必标定太多的摄像机的外参数 ,方法简洁 ,适合于视觉系统中对摄像机畸变的实时校正 ,或对摄像机捕获的图像进行几何校正 .实验表明 ,具有很强的鲁棒性和较高的校正精度     

借助大位移视图进行图像修补的透视畸变最小化算法

提出一种由粗到精的透视畸变最小化算法,借助大位移视图来修补目标图像.它先对大位移视点图像进行透视畸变校正后再用来补全目标图像上的丢失信息区域.首先,在平面场景的假设下,大位移视点图像通过单应矩阵进行全局变形得到初始的畸变校正.然后,由误匹配识别机制检测出初始校正的大位移视图中的残余畸变.在颜色一致性和位移场光滑性的期望下,残余畸变通过基于能量优化的重叠像素对应算法得到进一步的松弛.最后,在极线几何以及像素邻域中的位移场光滑性和颜色一致性的约束下,信息丢失像素按照特别定义的修补优先级函数依次得到恢复.泊松图像融合算法被用于消除修补区域与其周围像素之间可能存在的鬼影现象从而得到无缝的修补效果.实验表明,该方法优于已有的图像修补算法,且能够修补含有复杂结构信息的较大受损区域.     

基于目标跟踪的红外触摸屏优化算法

针对现有红外触摸屏多点识别时识别错乱和画图时轨迹毛刺严重等问题,在原有识别算法的基础上,加入了卡尔曼滤波与跟踪门算法结合的目标跟踪算法处理,提出一种新型红外触摸屏优化算法。该算法通过数据关联中的跟踪门逻辑,将量测值与已有的目标轨迹进行逻辑判断,选择正确的触点信息,剔除伪触点的信息,实现轨迹关联和目标跟踪。同时在红外触摸屏上建立触点的运动模型上,通过卡尔曼滤波实现轨迹的平滑和一定程度上的轨迹预测。在红外触摸屏上的测试表明,与原有的识别算法相比,优化算法增加了约3 μs单点识别时间,但是提高了轨迹在拐角处的平滑度,减少了拐角处约60%的毛刺。优化算法有效地修正了红外触摸屏原有的不足,能实现多触点时的轨迹关联和提高轨迹的平滑程度,能提高红外触摸屏绘图时的实际使用效果。     

多触点交互桌面左右手区分方法

现有多触点交互桌面系统通常只提供触点位置、形状等信息,不包括触点左右手归属信息.触点所属手及左右手手性信息的提供,对于多指手势识别、丰富双手交互技术,尤其是非对称双手交互技术具有重要意义.基于手的解剖结构特征,提出一种不需辅助硬件设备的触点左右手归属判定方法.首先,以手势设计的基本原则为指导,根据手的解剖结构特征,提出交互桌面手-臂系统三角形模型;其次,基于该三角形模型,给出多触点交互桌面同手触点聚类方法及左右手识别方法;然后,对MTDriver做扩展,实现了提供左右手信息的多触点跟踪工具箱,并提出了基于左右手信息的交互桌面交互技术;最后,评估结果表明,同手触点聚类方法及触点左右手手性识别方法均具有较高的正确率,而且在时间性能方面满足交互桌面交互实时性要求.     

Input Behavior When Using Two Fingers on a Multi-Touch Device

The purpose of this study is to analyze device input patterns using fingers in multi-touch mode and to derive Fitts’ law for multi-touch methods. Three specific input behaviors—dragging, rotating, and pinching—are investigated. Test participants performed two Fitts’ law experiments, a single-touch experiment using the thumb only and a multi-touch experiment using both the thumb and index finger. Within-subject factorial design was implemented with two levels of touch behavior (single-touch) and three levels of touch behavior (multi-touch), six levels of the index of difficulty (ID), four levels of direction, and two levels of repetition. The results indicate that a revised model of Fitts’ law is more appropriate for the multi-touch mode, achieving R² values above 0.895, and identifying the best nonlinear model among several regression models. The results of this study are used to determine the relationship between ID and movement time in terms of Fitts’ law and to investigate whether this relationship is sufficient to serve as a model to predict movement times. This study contributes to the prediction of movement times for products with multi-touch interfaces.     

OSG多点触控自然用户接口框架

OSG多点触控自然用户接口框架是在Windows多点触控技术基础上,将触控事件的管理和处理与OSG的事件处理机制相结合,形成了OSG的多点触控运行框架.在与用户交互过程中产生手指触屏的原始数据,根据这些原始数据定义所需的手势,并将其与OSG中交互事件处理机制相结合,完成利用多点触控对三维场景的交互操作.基于以上原理,分析了三维用户交互中主要的操作任务,定义了符合三维空间操作认知的多点触控交互手势,设计了相关手势的识别算法,并通过实例应用的开发验证了这一原理和设计成果的正确性和可行性.     

一种新型红外多点触摸识别算法

近年来,多点触摸技术逐渐成为最为重要的人机交互设备,得到了广泛的应用。本文提出一种新型红外多点触摸识别算法原型,该算法同时在两个不同坐标体系中对用户的触摸进行识别,克服了传统红外触摸屏在识别多点触摸时存在的伪触摸点的问题,实验证明该算法提高了红外触摸屏的多点触摸识别率。     

Stereometric pattern recognition by artificial touch

An artificial touch perception system has been created in order to study a method of processing information obtainable through tactile exploration of three-dimensional forms. The results can be useful for several purposes.The first part of the work concerns the project of the tactile explorator. For this purpose we used a kind of artificial limb like a finger with a certain number of touch sensitive transducers distributed along the surface of the finger tip. The information received by touching the object with the finger, is successively utilized as the input of the control servosystem which moves the finger point-by-point along the object surface in order to proceed with the exploration. It must be noticed that, from a philosophical point-of-view, the parallel approach with more fingers touching simultaneously the object in several points, is equal to sequential touching of these points by one moving finger.The second part describes the use of the propositional calculus in logical classification of the objects, as a method of three-dimensional pattern recognition.Elaboration of the input data obtained by tactile exploration, and computation of characteristic geometrical features of three-dimensional forms, has been performed on computer.     

Using technological functions on a multi-touch table and their affordances to counteract biases and foster collaborative problem solving

Touch technologies have become ubiquitous, motivating researchers to explore their potential - especially in collaborative scenarios. Studies on collaboration using joint visual spaces like multi-touch tables have demonstrated positive effects on performance. Yet, factors like prior knowledge and preferences, resulting in cognitive biases, were neglected although they are likely to put additional demands on collaboration. Whether touch technology can support its users in mastering the resulting challenges remains an open issue. To address this issue, we employed a hidden-profile paradigm (e.g., Schulz-Hardt and Mojzisch 2012) to investigate whether the affordances of specific support functions realized in a collaboration support kit on a multi-touch table help to overcome established pitfalls of collaboration (prior preferences and discussion biases). The collaboration support kit comprised a joint space and private spaces. It allowed participants to push information from the private into the joint space, to jointly sort information within the joint space, and it provided automatic functions like merging information. To replicate traditional hidden-profile studies, triads in a standard hidden-profile condition (n?=?25) exchanged information in a discussion; triads in the condition with collaboration support kit (n?=?29) were additionally provided with the aforementioned functions. Our results revealed that groups with collaboration support kit available showed greater discussion intensity, more balanced discussions, more indicators of mutual understanding, and better decision performance than standard hidden-profile groups. This is original evidence that affordances of a multi-touch table with interactive support functions can be used to overcome biases from prior preferences and to enhance collaboration.     

Establishing a baseline for text entry for a multi-touch virtual keyboard

Multi-touch, which has been heralded as a revolution in human–computer interaction, provides features such as gestural interaction, tangible interfaces, pen-based computing, and interface customization—features embraced by an increasingly tech-savvy public. However, multi-touch platforms have not been adopted as “everyday” computer interaction devices that support important text entry intensive applications such as word processing and spreadsheets. In this paper, we present two studies that begin to explore user performance and experience with entering text using a multi-touch input. The first study establishes a benchmark for text entry performance on a multi-touch platform across input modes that compare uppercase-only to mixed-case, single-touch to multi-touch and copy to memorization tasks. The second study includes mouse style interaction for formatting rich text to simulate a word processing task using multi-touch input. As expected, our results show that users do not perform as well in terms of text entry efficiency and speed using a multi-touch interface as with a traditional keyboard. Not as expected was the result that degradation in performance was significantly less for memorization versus copy tasks, and consequently willingness to use multi-touch was substantially higher (50% versus 26%) in the former case. Our results, which include preferred input styles of participants, also provide a baseline for further research to explore techniques for improving text entry performance on multi-touch systems.     

开源平台下的红外多点触摸屏驱动架构研究

随着计算机科学和人机交互技术的普及和发展,触摸技术起到了越来越重要的作用。本文主要介绍开源平台下红外多点触摸屏的特殊特点,并尝试研究一种红外多点触摸屏驱动架构,用以解决开源平台下的红外多点触摸屏驱动的应用问题。     

基于区域的手指三维运动跟踪

提出了基于区域的多连接体(手指)的三维运动跟踪算法.该算法首先用多约束融合的方法以及手指的运动特性,得到初始帧手指的三维结构;然后根据刚性多连接体的运动模型以及相应的姿势约束模型,给出了这一特殊运动模型三维运动估计的优化算法,此算法能够鲁棒地估计手指的三维运动;最后利用区域跟踪的方法获取多连接体三维运动,并在真实的手指序列图象中实现了该算法.实验结果证实了该算法的有效性.