Research on Three-Dimensional "High Presence Relay" System

三维“高临中继”系统研究

基本信息

批准号：
18500131
负责人：
WU Haiyuan
金额：
$ 2.58万
依托单位：
Wakayama University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2006
资助国家：
日本
起止时间：
2006 至 2007
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-18500131/
关键词：
High presence relay Video rate Target tracking Active camera Three-dimensional

项目摘要

A) In this research, we propose a K-means clustering based target tracking algorithm (called as K-means tracker), which can work robustly when tracking an object with hole through which the background can be seen (e.g. mosquito coil). The contributions of this research include: 1) Using a 5D feature vector to represent both the color "Y,U,V" and geometric "x,y" features of each pixel uniformly. This enables the simultaneous adaptation to both the color and geometric features during tracking; 2) Using a variable ellipse model to describe the search area and model the surrounding background. This guarantees the stable object tracking under various geometric transformations; 3) Achieving the automatic self tracking failure detection and recovery with both the "target" and "background" samples. Moreover, we propose a reliability-based K-means clustering algorithm (called as RK-means tracker) to remove the noise background pixel (which is neither similar to the target nor the background sam … More ples) from the target object. According to the triangular relationship among an unknown pixel and its two nearest cluster centers (target and background), the normal pixel (target or background one) will be assigned with high reliability value and correctly classified, while noise pixels will be given low reliability value and ignored. A radial sampling method is also brought out for improving both the processing speed and the robustness of this algorithm.B) In this research, we propose a high-performance object tracking system for obtaining high-quality images of a high-speed moving object at video rate by controlling a Fixed Viewpoint Pan Tilt Camera (FV-PTZ Camera) . To achieve our goal, we use the K-means tracker algorithm for tracking objects in an image sequence captured by the active cameras. We use the results of the K-means tracker to control the angular position and speed of each pan-tilt-zoom unit by employing the PID control scheme. Moreover, by using two cameras, the binocular stereo vision algorithm can be used to obtain the 3D position and velocity of the object. These results are used in order to adjust the focus and zoom. Our system allows the two cameras to gaze at a single point in 3D space. However, this system may become unstable when the time response deteriorates by excessively interfering in a mutual control loop or by strict restriction of the camera action. In order to solve these problems, we introduce the concept of reliability into the K-means tracker, and propose a method for controlling the active cameras by using relative reliability. Less

a）在这项研究中，我们提出了基于K-Means聚类的目标跟踪算法（称为K-Means Tracker），当可以看到可以看到背景的孔（例如蚊子线圈）时，该算法可以稳健地工作。这项研究的贡献包括：1）使用5D特征矢量代表每个像素的颜色“ y，u，v”和几何“ x，y”特征均匀。这使得在跟踪过程中同时适应了颜色和几何特征。 2）使用可变椭圆模型来描述搜索区域并建模周围背景。这样可以保证在各种几何变换下稳定的对象跟踪； 3）使用“目标”和“背景”样本实现自动自动跟踪失败检测和恢复。此外，我们提出了一种基于可靠性的K-均值聚类算法（称为RK-Means Tracker），以从目标对象中删除噪声背景像素（既不类似于目标，也不类似于目标SAM…更多的PLES）。根据未知像素及其最近的两个群集中心（目标和背景）之间的三角关系，将分配具有高可靠性值并正确分类的正常像素（目标或背景一个），而噪声像素将被忽略较低并忽略。在这项研究中，还提出了一种径向采样方法，以提高该算法的处理速度和稳健性。为了实现我们的目标，我们使用K-Means跟踪器算法在活动摄像机捕获的图像序列中跟踪对象。我们使用K-均值跟踪器的结果来控制每个板倾斜单元的角度和速度，并采用PID控制方案。此外，通过使用两个摄像机，可以使用双眼立体视觉算法来获得物体的3D位置和速度。这些结果用于调整焦点和变焦。我们的系统允许两个摄像机凝视3D空间的一个点。但是，当时间响应通过极大干扰相互控制循环或严格限制相机动作来检测时，该系统可能会变得不稳定。为了解决这些问题，我们将可靠性概念介绍到K-均值跟踪器中，并提出了一种通过使用相对可靠性来控制活动摄像机的方法。较少的