基于主动视觉和超声的机器人目标跟踪系统

运动目标跟踪技术是未知环境下移动机器人研究领域的一个重要研究方向。该文提出了一种基于主动视觉和超声信息的移动机器人运动目标跟踪设计方法,利用一台SONY EV-D31彩色摄像机、自主研制的摄像机控制模块、图像采集与处理单元等构建了主动视觉系统。移动机器人采用了基于行为的分布式控制体系结构,利用主动视觉锁定运动目标,通过超声系统感知外部环境信息,能在未知的、动态的、非结构化复杂环境中可靠地跟踪运动目标。实验表明机器人具有较高的鲁棒性,运动目标跟踪系统运行可靠。

Global Translation Preferred Active Contour

具有全局平移优先属性的主动轮廓更适于目标跟踪。演化轮廓具有的全局平移优先性可以理解为沿轮廓的速度场具有相等的倾向。根据此思想,通过定义在曲线扰动集合上的新内积空间导出了一种简单,具有平移优先的梯度流。新的内积空间由于是通过向H0主动轮廓对应的內积空间引入曲线扰动的方差获得,所以此主动轮廓称为方差主动轮廓。方差主动轮廓是将H0主动轮廓与其对应的平均梯度流通过加权求和获得,而H1主动轮廓则是通过H0主动轮廓与特定类型的核函数进行卷积得到。因此方差主动轮廓实现时更简单和快速。最后给出了H0,H1和方差主动轮廓在频域与时域的分析。

移动机器人运动目标跟踪系统设计

利用SONYEV-D31摄像机和自主研发的摄像机控制模块,构建了一套主动视觉子系统,并将该子系统应用于RIRA-II型移动机器人上,实现了移动机器人运动目标自动跟踪功能。RIRA-II移动机器人采用了由一组分布式行为模块和集中命令仲裁器组成的基于行为的分布式控制体系结构。各行为模块基于领域知识通过反应方式产生投票,由仲裁器产生动作指令,机器人完成相应的动作。在设置了障碍、窄通道以及模拟墙体的复杂环境下进行运动目标跟踪实验,实验表明运动目标跟踪系统运行可靠,具有较高的鲁棒性。

基于多微处理器的智能超声探测系统

设计了一种基于多微处理器的智能超声探测系统。该系统采用由上位工作模式控制单元和下位智能超声传感器阵列组成的分布式结构。每个超声传感器均由独立的微处理器控制,并完成实时数据处理、抗干扰处理、故障报警以及数据通信等功能。上位控制单元根据移动机器人的运动状态采用不同的控制策略,使下位传感器阵列分组并行工作,提高了探测系统的实时性;下位传感器采用"阈值比较法"、"改进型递推均值滤波"算法和模糊信息处理技术,提高了探测系统的准确性、可靠性。将该探测系统安装于RIRA-II移动机器人上,进行了基于主动视觉和超声信息的运动目标跟踪实验,实验表明超声探测系统运行可靠、稳定。

空间工作记忆与注意广度

The relationship between working memory (WM) and attention has attracted many researchers. In his embedded-processes model of WM, Cowan (1999, 2001) uses the term focus of attention to refer to the core component of WM, and proposes that the focus of attention of WM and that of perception have the same span, which is a fixed number. This hypothesis about the scope of attention has seldom been tested, although considerable studies have revealed that WM and attention have overlapping mechanisms. The present dissertation tests this hypothesis by examining the dual-task interference between Corsi Blocks Task (CBT) and Mutilple Object Tracking (MOT) and the findings demenstrate that Cowan’s hypothesis is not exactly true. The results of our first study show that the interference effect of MOT on CBT is a reliable indicator of whether and to which extent the attentional resoureces of WM and perception overlap. In the second study we find that the capacity of the common resources is not a fixed number but varies with the difficulty of control of attention. And the third study indicates that attentional resources used in WM and perception are partly independent, the overlapping part can attend to only one or two items or locations at a time. These findings can contribute to future studies on the capacity limitation of different cognitive functions, and to the development of relevant ability tests.

Multi-scale keypoint annotation: a biological approach

The primary visual cortex employs simple, complex and end-stopped cells to create a scale space of 1D singularities (lines and edges) and of 2D singularities (line and edge junctions and crossings called keypoints). In this paper we show first results of a biological model which attributes information of the local image structure to keypoints at all scales, ie junction type (L, T, +) and main line/edge orientations. Keypoint annotation in combination with coarse to fine scale processing facilitates various processes, such as image matching (stereo and optical flow), object segregation and object tracking.

Automatic Video Analysis for Fisheries Survey Systems

Thesis (Ph.D.)--University of Washington, 2015

Mouvement biologique et entraînement perceptivo-cognitif chez les personnes âgées

Lorsque nous cherchons un ami dans une foule ou attendons un proche sur le quai d’une gare, l’identification de cette personne nous est souvent possible grâce à la reconnaissance de sa démarche. Plusieurs chercheurs se sont intéressés à la façon de se mouvoir de l’être humain en étudiant le mouvement biologique. Le mouvement biologique est la représentation, par un ensemble structuré de points lumineux animés, des gestes d’un individu en mouvement dans une situation particulière (marche, golf, tennis, etc.). Une des caractéristiques du patron de mouvement biologique peu étudiée et néanmoins essentielle est sa taille. La plupart des études concernées utilisent des patrons de petite taille correspondant à une personne située à 16 mètres de l’observateur. Or les distances d’interaction sociale, chez l’humain, sont généralement inférieures à 16 mètres. D’autre part, les résultats des études portant sur la perception des patrons de mouvement biologique et le vieillissement demeurent contradictoires. Nous avons donc, dans un premier temps, évalué, dans une voûte d’immersion en réalité virtuelle, l’importance de la distance entre l’observateur et le patron de mouvement biologique, chez des adultes jeunes et des personnes âgées. Cette étude a démontré que l’évaluation de la direction de mouvement d’un patron devient difficile pour les personnes âgées lorsque le patron est situé à moins de 4 mètres, alors que les résultats des jeunes sont comparables pour toutes distances, à partir d’un mètre et au-delà. Cela indique que les gens âgés peinent à intégrer l’information occupant une portion étendue de leur champ visuel, ce qui peut s’avérer problématique dans des espaces où les distances d’interaction sont inférieures à 4 mètres. Nombre de recherches indiquent aussi clairement que les gens âgés s’adaptent difficilement à des situations complexes. Nous avons donc cherché, dans un second temps, à minimiser ces altérations liées à l’âge de l’intégration des processus complexes, en utilisant une tâche adaptée à l’entraînement et à l’évaluation de l’intégration de ces processus : la poursuite multiple d’objets dans l’espace ou 3D-MOT (3 Dimensions Multiple Object Tracking). Le 3D-MOT consiste à suivre simultanément plusieurs objets d’intérêt en mouvement parmi des distracteurs également en mouvement. Nous avons évalué les habiletés de participants jeunes et âgés à une telle tâche dans un environnement virtuel en 3D en déterminant la vitesse maximale de déplacement des objets à laquelle la tâche pouvait être exécutée. Les résultats des participants âgés étaient initialement inférieurs à ceux des jeunes. Cependant, après plusieurs semaines d’entraînement, les personnes âgées ont obtenu des résultats comparables à ceux des sujets jeunes non entraînés. Nous avons enfin évalué, pour ces mêmes participants, l’impact de cet entraînement sur la perception de patrons de mouvement biologique présentés à 4 et 16 mètres dans l’espace virtuel : les habiletés des personnes âgées entraînées obtenues à 4 mètres ont augmenté de façon significative pour atteindre le niveau de celles obtenues à 16 mètres. Ces résultats suggèrent que l’entraînement à certaines tâches peut réduire les déclins cognitivo-perceptifs liés à l’âge et possiblement aider les personnes âgées dans leurs déplacements quotidiens.

3D path planning with novel multiple 2D layered approach for complex human-robot interaction

Navigating cluttered indoor environments is a difficult problem in indoor service robotics. The Acroboter concept, a novel approach to indoor locomotion, represents unique opportunity to avoid obstacles in indoor environments by navigating the ceiling plane. This mode of locomotion requires the ability to accurately detect obstacles, and plan 3D trajectories through the environment. This paper presents the development of a resilient object tracking system, as well as a novel approach to generating 3D paths suitable for such robot configurations. Distributed human-machine interfacing allowing simulation previewing of actions is also considered in the developed system architecture.

Methods of sensory augmentation through electro-tactile stimulation of the tongue

This paper details an investigation into sensory substitution by means of direct electrical stimulation of the tongue for the purpose of information input to the human brain. In particular, a device has been constructed and a series of trials have been performed in order to demonstrate the efficacy and performance of an electro-tactile array mounted onto the tongue surface for the purpose of sensory augmentation. Tests have shown that by using a low resolution array a computer-human feedback loop can be successfully implemented by humans in order to complete tasks such as object tracking, surface shape identification and shape recognition with no training or prior experience with the device. Comparisons of this technique have been made with visual alternatives and these show that the tongue based tactile array can match such methods in convenience and accuracy in performing simple tasks.

Robust video/ultrasonic fusion-based estimation for automotive applications

In this paper, we use recently developed robust estimation ideas to improve object tracking by a stationary or nonstationary camera. Large uncertainties are always present in vision-based systems, particularly, in relation to the estimation of the initial state as well as the measurement of object motion. The robustness of these systems can be significantly improved by employing a robust extended Kalman filter (REKF). The system performance can also be enhanced by increasing the spatial diversity in measurements via employing additional cameras for video capture. We compare the performances of various image segmentation techniques in moving-object localization and show that normal-flow-based segmentation yields comparable results to, but requires significantly less time than, optical-flow-based segmentation. We also demonstrate with simulations that dynamic system modeling coupled with the application of an REKF significantly improves the estimation system performance, particularly, when subjected to large uncertainties.

tCAD: a 3D modeling application on a depth enhanced tabletop computer

Tabletop computers featuring multi-touch input and object tracking are a common platform for research on Tangible User Interfaces (also known as Tangible Interaction). However, such systems are confined to sensing activity on the tabletop surface, disregarding the rich and relatively unexplored interaction canvas above the tabletop. This dissertation contributes with tCAD, a 3D modeling tool combining fiducial marker tracking, finger tracking and depth sensing in a single system. This dissertation presents the technical details of how these features were integrated, attesting to its viability through the design, development and early evaluation of the tCAD application. A key aspect of this work is a description of the interaction techniques enabled by merging tracked objects with direct user input on and above a table surface.

Increasing Realism and Supporting Content Planning for Dynamic Scenes in a Mixed Reality System incorporating a Time-of-Flight Camera

For broadcasting purposes MIXED REALITY, the combination of real and virtual scene content, has become ubiquitous nowadays. Mixed Reality recording still requires expensive studio setups and is often limited to simple color keying. We present a system for Mixed Reality applications which uses depth keying and provides threedimensional mixing of real and artificial content. It features enhanced realism through automatic shadow computation which we consider a core issue to obtain realism and a convincing visual perception, besides the correct alignment of the two modalities and correct occlusion handling. Furthermore we present a possibility to support placement of virtual content in the scene. Core feature of our system is the incorporation of a TIME-OF-FLIGHT (TOF)-camera device. This device delivers real-time depth images of the environment at a reasonable resolution and quality. This camera is used to build a static environment model and it also allows correct handling of mutual occlusions between real and virtual content, shadow computation and enhanced content planning. The presented system is inexpensive, compact, mobile, flexible and provides convenient calibration procedures. Chroma-keying is replaced by depth-keying which is efficiently performed on the GRAPHICS PROCESSING UNIT (GPU) by the usage of an environment model and the current ToF-camera image. Automatic extraction and tracking of dynamic scene content is herewith performed and this information is used for planning and alignment of virtual content. An additional sustainable feature is that depth maps of the mixed content are available in real-time, which makes the approach suitable for future 3DTV productions. The presented paper gives an overview of the whole system approach including camera calibration, environment model generation, real-time keying and mixing of virtual and real content, shadowing for virtual content and dynamic object tracking for content planning.