An active contour model for object tracking using the previous contour

An active contour model is proposed for object tracking using prior information. Conventional algorithms have many problems when applied in object tracking. The proposed active contour algorithm, a model using an edge of an adapted color feature, not only modifies the internal energy function of the conventional algorithm to extend the search range and reduce the computational burden, but also modifies the external energy function to reduce the edge candidates of the object. The algorithm searches normally and uses dynamic programming to solve the energy minimization problem. The main drawbacks of a conventional snake algorithm, i.e., shrinking, a limited search range, sensitivity to outliers, are improved with the proposed algorithm. We illustrate the effectiveness of our scheme using some tracking examples. This work was presented, in part, at the Seventh International Symposium on Artificial Life and Robotics, Oita, Japan, January 16–18, 2002     

基于双向轮廓跟踪的面部轮廓线生成算法

为了提高医学图像中面部轮廓线的分割效果,提出一种新的面部轮廓线生成算法(bidirectional contour tracking,BCT).该算法首先利用边缘检测算法对CT图像进行检测,然后利用头部CT图像中各组织的灰度特征和图像的对称性,对检测后的图像进行双向轮廓跟踪,并获得面部的轮廓线.另外,根据面部轮廓线的特点,使用中点法可生成辅助轮廓线.以一组颅脑CT图像为例来验证算法的效果,实验结果表明,新的生成算法提高了轮廓线的生成速度,并有很好的分割效果.     

活动轮廓模型目标跟踪算法综述

目标跟踪是当前计算机视觉领域最活跃的研究主题。首先对基本的跟踪类型进行了介绍;然后讨论了基于活动轮廓模型的图像分割,重点分析了参数活动轮廓模型的梯度矢量流模型（Gradient Vector Flow,GVF）,以及几何活动轮廓模型中的模型;并讨论了基于粒子滤波的目标跟踪算法的研究现状,最后展望了这一领域未来研究的热点。     

Efficient edge-based object tracking

Object tracking has been widely applied to video surveillance, robot localization and human-computer interaction. In this paper, an edge-based tracking algorithm is proposed. We extract the feature points by efficiently utilizing the image edges in the object region. Then the parameter vector of the object's motion model is estimated based on minimizing the sum-of-squared differences between the reference feature points in the reference frame and the observed feature points in the tracking sequence frame. The experiments show that the edge-based tracking algorithm proposed by us can track object efficiently under uniform and varying illumination conditions.     

基于边缘方向分布和粒子滤波的视频目标跟踪

针对背景复杂的非线性视频运动目标,提出了一种基于边缘方向分布和粒子滤波技术相结合的跟踪方法。该方法利用边缘算子获取目标区域边缘方向图,采用高斯核函数建立目标区域边缘方向分布,结合粒子滤波理论,实现对非线性视频运动目标的有效跟踪。计算机仿真结果表明,该方法可对非线性、非高斯的运动目标进行有效的跟踪,且在跟踪目标区域与背景颜色接近,背景复杂的场景下,与采用灰度特征的方法相比,有较强的鲁棒性和较高的跟踪精度。     

基于置信度策略选择的实时目标跟踪方法

目标跟踪一直是计算机视觉领域研究的热点和难点,受自然场景中复杂干扰因素影响,现有方法的速度和精度尚待改善。本文首先对基于颜色属性的目标跟踪算法改进,使之更为鲁棒且速度达到实时。接下来,针对被跟踪目标发生遮挡时,采用基于颜色属性的跟踪算法导致错误累积进而产生漂移甚至跟踪失败的问题,引入运算量较大但对遮挡有较强抵抗能力的稀疏协作表观模型。为了同时保证算法的速度和准确性,本文构建了一套基于跟踪结果置信度评量的策略选择机制,将两种算法有机整合。在多个公开数据集下的对比实验显示,与现有跟踪算法相比,本文方法在跟踪效果和速度上具有较显著优势,并在目标存在严重遮挡、光照变化、运动模糊等情况时,均可以取得较好的跟踪效果。     

Usability of various dwell times for eye-gaze-based object selection with eye tracking

This study investigates the usability of various “dwell times” for selecting visual objects with eye-gaze-based input by means of eye tracking. Two experiments are described in which participants used eye-gaze-based input to select visual objects consisting of alphanumeric characters, dots, or visual icons. First, a preliminary experiment was designed to identify the range of dwell time durations suitable for eye-gaze-based object selection. Twelve participants were asked to evaluate, on a 7-point rating scale, how easily they could perform an object-selection task with a dwell time of 250, 500, 1000, or 2000 ms per object. The evaluations showed that a dwell time of 250 ms to around 1000 ms was rated as potentially useful for object selection with eye-gaze-based input. In the following main experiment, therefore, 30 participants used eye tracking to select object sequences from a display with a dwell time of 200, 400, 800, 1000 or 1200 ms per object. Object selection time, object selection success rate, the number of object selection corrections, and dwell time evaluations were obtained. The results showed that the total time necessary to select visual objects (object selection time) increased when dwell time increased, but longer dwell times resulted in a higher object-selection success rate and fewer object selection corrections. Furthermore, regardless of object type, eye-gaze-based object selection with dwell times of 200–800 ms was significantly slower for participants with glasses than for those without glasses. Most importantly, participant evaluations showed that a dwell time of 600 ms per object was easiest to use for eye-gaze-based selection of all three types of visual objects.     

基于图像轮廓检测的航天器目标跟踪控制系统设计

目前研究的航天器目标跟踪控制系统控制有效率低,追踪图像与实际目标不同,准确率低;基于图像轮廓检测设计了一种新的航天器目标跟踪控制系统,根据系统硬件的不同性能与结构进行模块式划分,同时将图像轮廓检测数据准则添加入中心管理系统中,时刻保证数据的检测安全系数处于系统操作允许范围内;调整原有的软件结构状态,采集航天器轮廓图像并提取轮廓信息,选用适当的图像轮廓检测阈值,集中检验跟踪目标的运动方向,调控方向数据,以此降低图像的像素变化程度,得到准确的航天器目标跟踪控制结果;实验结果表明基于图像轮廓检测的航天器目标跟踪控制系统控制有效率达到了84%,跟踪目标更加准确。     

Scalable continuous object detection and tracking in sensor networks

With the advancement of MEMS technologies, sensor networks have opened up broad application prospects. An important issue in wireless sensor networks is object detection and tracking, which typically involves two basic components, collaborative data processing and object location reporting. The former aims to have sensors collaborating in determining a concise digest of object location information, while the latter aims to transport a concise digest to sink in a timely manner. This issue has been intensively studied in individual objects, such as intruders. However, the characteristic of continuous objects has posed new challenges to this issue. Continuous objects can diffuse, increase in size, or split into multiple continuous objects, such as a noxious gas. In this paper, a scalable, topology-control-based approach for continuous object detection and tracking is proposed. Extensive simulations are conducted, which show a significant improvement over existing solutions.     

基于边缘直方图的粒子滤波跟踪算法及其实现

以颜色直方图为特征的运动目标跟踪算法容易受到光线变化及视场内其它同色目标的干扰.采用运动目标的边缘方向直方图作为特征,利用序列重要性采样原理和粒子滤波算法实现了对人体运动目标的跟踪.实验显示了该算法在光线变化及存在同色目标干扰时能够有效跟踪目标.在算法实现过程中,采用积分图计算边缘方向直方图,减少了计算时间,提高了计算速度,达到了实时跟踪的效果.     

基于Snake模型的视频对象分割和跟踪算法

视频对象的分割是基于内容的视频处理中重要的组成部分。提出并实现了一种半自动视频对象分割和跟踪算法。算法主要基于Williams活动轮廓模型，通过求取轮廓点的局部能量最小值对轮廓线进行更新。轮廓扩张技术用来追踪变形的轮廓边缘。通过对轮廓中心点运动的统计，预测对象的运动方向和大小。实验仿真结果表明，这种改进的Snake算法能够收缩到图像的凹陷部分，而且能较好地跟踪视频对象的运动。     

采用测地线活动轮廓模型检测与跟踪运动目标

水平集几何活动轮廓模型能较好地适应曲线的拓扑变化.为了跟踪和获取刚体和非刚体运动目标的轮廓信息,提出了一种基于改进测地线活动轮廓(GAC)模型和Kalman滤波相结合的算法以检测和跟踪运动目标.该算法首先采用高斯混合模型和背景差分获取目标的运动区域,在运动区域内采用引入距离规则化项的GAC模型进行曲线演化,使改进GAC模型在运动目标的真实轮廓处收敛;然后通过结合Kalman滤波预测目标下一帧的位置,实现对目标轮廓跟踪.实验结果表明,该方法适用于刚体和非刚体目标,在部分遮挡的情况下也能保持良好的检测和跟踪效果.     

联合轮廓法在低分辨率视频下的多目标追踪

相似目标间的遮挡一直是多目标追踪问题中的难点。针对这个问题提出了联合轮廓法。这种方法在使用粒子滤波器的基础上,通过对遮挡目标集的外轮廓建模来确定处于遮挡关系中目标的状态,从而进一步得到目标的轨迹。算法在实验中显示了良好的鲁棒性,并且计算复杂度与遮挡目标的个数成二次多项式关系。     

新颖的物体轮廓提取：基于双边滤波的多级边缘检测*

提出了一种新颖的物体轮廓提取方法,即通过多级边缘检测来提取物体的主要边界。双边滤波器用来建立多级,同时Canny边缘算子相应地产生边缘图,组合边缘图构造出一幅多级图。次段被定义为边缘像素的连接结构,提取并连接这些次段可以构成闭合轮廓。最终图像中最相关的闭合轮廓被判定为真实的物体轮廓。实验结果表明,该物体轮廓提取方法具有较高的可靠性并且受噪声影响较小。     

一种基于脑血管图像特征的快速边界跟踪算法*

在处理复杂的脑血管图像时,经典的边界跟踪算法暴露出边界精度不高、边界不够平滑,且速度不尽人意等缺点。提出了一种新的快速边界跟踪算法,该算法在分析脑血管边缘垂直细节远多于水平细节的特征基础上,结合方向记忆选择搜索方向,并在不同的搜索方向上赋予不同的权值,最终得到下一个边界点。实验表明：该算法提取的脑血管边界平滑、速度快,适合脑血管图像的边界提取,为下一步的脑血管形状特征提取及表示提供了精确的数据准备。     

基于视觉注意机制的目标跟踪方法

提出一种基于视觉注意机制的运动目标跟踪方法。该方法借鉴人类的视觉注意机制的研究成果,建立视觉注意机制的计算模型,计算视频中各部分内容的视觉显著性。结合视觉显著性计算结果,提取视频图像中的显著性目标。利用颜色分布模型作为目标的特征表示模型,与视频中各显著目标进行特征匹配,实现目标的跟踪。在多个视频序列中进行实验,并给出相应的实验结果及分析。实验结果表明,提出的目标检测与跟踪算法是正确有效的。     

一种基于粒子群优化的目标跟踪特征选择算法

针对复杂背景下的运动目标跟踪特征选择问题,提出了一种基于粒子群优化的目标跟踪特征选择算法。假设具有目标与背景间最好可分离性的特征为最好的跟踪特征。通过构建目标与背景的图像特征分布方差的比值函数作为衡量目标与背景间的可分离性判据。使用粒子群优化算法优化不同的特征组合实时获取最优的目标跟踪特征。为验证该算法的有效性,将选择的最优特征与一种基于核的跟踪算法相结合进行跟踪实验。实验结果表明,算法能有效提高传统基于核的跟踪算法对于复杂场景下的运动目标跟踪的鲁棒性与准确性。     

基于全局最小化活动轮廓的多目标检测跟踪

为了在噪声干扰以及目标和背景颜色相近情况下实现多目标跟踪,提出一种基于快速全局最小化的活动轮廓模型的目标检测跟踪算法。该算法结合了基于边缘的活动轮廓模型和基于区域的活动轮廓模型,对能量泛函进行全局最小化来检测目标活动轮廓,用卡尔曼滤波预测目标下一帧的特征信息,然后用改进的最近邻法进行多目标跟踪。对图像序列的实验结果表明该算法能有效地对运动背景下多目标进行跟踪。