基于特征融合及自适应模型更新的相关滤波目标跟踪算法

针对单一特征目标跟踪算法因背景干扰、目标遮挡造成的跟踪失败问题,以及跟踪过程中每帧进行模型更新容易造成错误更新和实时性差的问题,提出了一种基于特征融合及自适应模型更新策略的相关滤波目标跟踪算法-多特征自适应相关滤波目标跟踪算法。该算法在特征提取阶段将边缘特征及HOG特征加权融合作为目标特征,加强对边缘特征的学习;在模型更新阶段通过计算预测区域与真实区域的奇异值特征向量相似度,并结合设定的阈值判断是否需要进行模型更新,通过自适应更新的方式减少模型的更新次数。在标准测试视频集下验证所提算法,并与两种经典相关滤波算法进行比较,结果表明该算法能够较好地适应背景干扰及目标遮挡问题,跟踪目标的平均中心误差减少了9.05像素,平均距离精度提高12.2%,平均重叠率提高4.53%。     

基于混合高斯模型的Mean-Shift模板更新算法

针对基于Mean-Shift目标跟踪算法中遇到的不能对模板进行实时更新的问题,提出一个基于混合高斯背景建模的目标模板更新算法.该算法将目标视为背景,对目标中的每一个像素点利用三个高斯函数对它进行建模,利用每次Mean-Shift跟踪到的目标区域来对先前建立的混合高斯模型进行实时更新,将混合高斯模型得到的目标模板作为下一帧跟踪的目标模板.该算法较好地解决了基于Mean-Shift算法的模板更新问题,实验证明该算法是有效的.     

基于分块背景更新的均值飘移跟踪算法

为提高融合背景信息均值飘移跟踪算法中的背景模型更新效率,提出基于分块背景更新的改进算法。将目标周围的背景区域划分成4块,分别以此4个子块的颜色直方图作为区域特征描述符进行计算,利用巴氏系数选择性地更新相似性较低的3个区域。在构建目标候选模型时,利用目标物体空间信息赋予候选目标区域像素点相应的距离权重。对整块背景信息更新策略和加权更新背景信息策略进行比较,比较结果表明,该算法具有更高的迭代收敛效率和更强的抗背景干扰能力。     

适用于遮挡问题的目标跟踪算法

提出一种基于网格模型的目标跟踪算法．该算法首先进行遮挡区域检测，然后进行网格结点的运动估计和网格更新过程完成目标的多帧跟踪．改进的遮挡区域检测算法有效地提高了检测准确度，从而确保遮挡区域的准确跟踪；网格结点的运动估计是通过特征窗口运动补偿匹配完成，可以有效地克服块效应．实验证明，该算法解决了二维运动估计时网格模型在遮挡区域存在的问题，并可以有效地进行目标准确跟踪．     

基于自适应Kalman滤波的机器人运动目标跟踪算法

针对足球机器人比赛时的模型变化及其环境噪声先验估计不准确的问题,提出一种基于自适应卡尔曼滤波的足球机器人视觉跟踪算法;该算法将一种基于减背景的运动目标识别的方法与自适应卡尔曼滤波跟踪模型进行结合,对背景进行实时更新,并通过形态学滤波去除残留的小区域,从而准确的识别运动目标,通过自适应的在线调整运动模型参数来保证模型预测值的准确性,进而提高了目标跟踪时的匹配效率,实现了目标的精准、迅速跟踪;通过实验证明,该算法是很有效的,具有推广价值.     

一种基于标记修正和样本更新的目标跟踪算法

针对半监督自训练框架下进行目标跟踪的误差累积问题,提出一种结合标记修正与区域置信度样本更新的自适应跟踪算法。该算法将视频序列中的目标跟踪视为两类模式即目标与背景的分类问题,在半监督自训练框架下,选择SVM分类器分类目标与背景,结合K近邻和最小距离分类进行标记修正,并基于区域置信度提取新的样本更新分类器。实验结果显示,该方法有效改善了由于误差累积导致的漂移问题和目标遮挡后的跟踪失败。     

利用背景加权和选择性子模型更新的视觉跟踪算法

目的 针对传统Mean Shift算法对受背景干扰的目标无法进行有效跟踪并缺少有效的模型更新策略的问题,提出一种将背景加权和选择性子模型更新相结合的跟踪算法。方法 首先,在Mean Shift框架下,为了减少背景信息对目标定位的干扰,利用目标区域周围像素的颜色直方图定义背景加权系数,并将该系数只引入到目标模型的颜色直方图中,从而建立一个新的目标模型。然后,根据目标模型中每个分量匹配贡献度的大小选取需要更新的模型分量及其更新公式。结果 实验结果表明,本文算法能够抑制背景干扰,同时能对模型进行有效的选择性更新,克服了整体更新策略严重的模型漂移问题。结论 本文从模型描述和更新策略两个方面对传统Mean Shift算法进行了改进,实验结果表明本文算法具有较好的有效性和鲁棒性。     

基于区域生长的Mean shift动态变形手势跟踪算法

针对传统Mean shift算法在手势跟踪中由于搜索窗口内手势背景像素改变所导致的跟踪精度不高,以及算法本身由于手势模型更新所引起的时间消耗较多等问题,提出一种基于区域生长与Mean shift算法相结合的动态变形手势跟踪算法.该算法在跟踪初始阶段通过帧间差分法对手势中心完成自动初始化,利用区域生长算法采集手势样本点,并通过Mean shift算法对目标中心进行精确定位.实验结果表明,该方法能够对动态变形手势实现精确实时的跟踪,可较好地降低算法的时间复杂度,保证运动目标跟踪的稳定性和连续性.     

基于加权颜色直方图和粒子滤波的彩色物体跟踪

结合粒子滤波技术，提出一种基于加权颜色直方图的彩色物体跟踪算法．将目标颜色直方图作为目标的颜色模型，同时考虑其大小及像素点的位置对颜色分布的影响，将颜色直方图进行加权处理，使模型对区域特征描述更加合理，利用巴特查理亚距离描述粒子与目标颜色模型的相似性，作为粒子更新权值的有力依据．目标颜色模型的合理建立使得算法的粒子需求量少，计算复杂度降低，利于实现实时跟踪．试验结果验证了该算法的有效性和实用性。     

基于多模态背景模型和霍夫森林的红外目标跟踪

针对红外目标跟踪过程中目标和背景耦合性强和目标形变带来的表观特征变化的问题,提出了基于多模态背景模型和霍夫森林的红外目标跟踪算法,将目标和背景的特征信息整体融入跟踪框架,并引入霍夫森林对目标部件表征信息进行训练更新;通过特征描述和模式列表建立多模态背景模型,在观测序列中记录匹配的背景模式列表,根据匹配结果进行权重和模态参数更新;以背景建模获得的区域信息为采样基础,将前景区域和背景区域分别作为正负样本集代入决策树进行训练;根据叶节点存储的图像块信息对目标位置进行投票决策,从图像特征空间映射到霍夫参数空间,生成目标区域概率分布图,获取高置信度区作为目标跟踪区。在红外公开测试集上实验结果表明,在背景杂波和目标形变对跟踪造成干扰的情况下,所提算法仍能保持跟踪的稳定性。     

Erosion band signatures for spatial extraction of features

We introduce Erosion Band Signatures (EBS), which are a codification of the spatial coherence of features extracted from a region. This coherence is often lost in traditional global and local feature extraction methods, thereby diminishing a feature’s discriminative strength. The erosion band signature is generated through iterative erosions of the region of interest, forming what we call erosion bands. Features are then extracted from each band and accumulated in a specific order to form the EB signature, which preserves spatial information of the features. To demonstrate the versatility of EBS, we have implemented the method in two very different applications: polyp detection and region-based head tracking. In polyp detection, EBS provides an effective way to characterize spatial differences between the perimeter and core of a polyp candidate, and improves a state-of-the-art computer-aided detection method with an improved 27.6% reduction of false positives. We also apply EBS analysis to region-based tracking yielding a very clear improvement in both robustness and accuracy.     

结合全局与局部信息活动轮廓的非同质图像分割

目的 通过对现有基于区域的活动轮廓模型能量泛函的Euler-Lagrange方程进行变形,建立其与K-means方法的等价关系,提出一种新的基于K-means活动轮廓模型,该模型能有效分割灰度非同质图像。方法 结合图像全局和局部信息,根据交互熵的特性,提出新的局部自适应权重,它根据像素点所在邻域的局部统计信息自适应地确定各个像素点的分割阈值,排除灰度非同质分割目标的影响。结果 采用Jaccard相似系数-JS(Jaccard similarity)和Dice相似系数-DSC(Dice similarity coefficient)两个指标对自然以及合成图像的分割结果进行定量分析,与传统及最新经典的活动轮廓模型相比,新模型JS和DSC的值最接近1,且迭代次数不多于50次。提出的模型具有较高的计算效率和准确率。结论 通过大量实验发现,新模型结合图像全局和局部信息,利用交互熵特性得到自适应权重,对初始曲线位置具有稳定性,且对灰度非同质图像具有较好地分割效果。本文算法主要适用于分割含有噪声及灰度非同质的医学图像,而且分割结果对初始轮廓具有鲁棒性。     

Salient object detection based on regions

Salient object detection aims to automatically localize the attractive objects with respect to surrounding background in an image. It can be applied to image browsing, image cropping, image compression, content-based image retrieval, and etc. In the literature, the low-level (pixel-based) features (e.g., color and gradient) were usually adopted for modeling and computing visual attention; these methods are straightforward and efficient but limited by performance, due to losing global organization and inference. Some recent works attempt to use the region-based features but often lead to incomplete object detection. In this paper, we propose an efficient approach of salient object detection using region-based representation, in which two novel region-based features are extracted for proposing salient map and the salient object are localized with a region growing algorithm. Its brief procedure includes: 1) image segmentation to get disjoint regions with characteristic consistency; 2) region clustering; 3) computation of the region-based center-surround feature and color-distribution feature; 4) combination of the two features to propose the saliency map; 5) region growing for detecting salient object. In the experiments, we evaluate our method with the public dataset provided by Microsoft Research Asia. The experimental results show that the new approach outperforms other four state-of-the-arts methods with regard to precision, recall and F-measure.     

一种快速的自适应目标跟踪方法

由于光照变化、视角差异、相机抖动和部分遮挡等因素的影响,鲁棒的目标跟踪仍然是计算机视觉领域极具挑战性的研究课题.受协同训练和粒子滤波算法的启发,提出一种快速的自适应目标跟踪方法.该方法采用HOG(histogram of oriented gradients)和LBP(local binary pattern)描述目标特征并建立分类器,通过协同训练实现分类器的在线更新,有效解决了误差累积问题.为缩小目标搜索的状态空间,利用ICONDENSATION的运动模型和重要采样提高粒子采样的准确性和效率,并引入校正因子抑制虚假目标的干扰,从而提升了跟踪算法的鲁棒性和分类器更新的准确性.在两组标准测试集和两组自建测试集上的对比实验结果验证了所提出跟踪算法的有效性.与基于全局搜索的跟踪方法相比,该算法在不降低跟踪性能的前提下将处理速度提高25倍以上.     

Fast object tracking using adaptive block matching

We propose a fast object tracking algorithm that predicts the object contour using motion vector information. The segmentation step common in region-based tracking methods is avoided, except for the initialization of the object. Tracking is achieved by predicting the object boundary using block motion vectors followed by updating the contour using occlusions/disocclusion detection. An adaptive block-based approach has been used for estimating motion between frames. An efficient modulation scheme is used to control the gap between frames used for motion estimation. The algorithm for detecting disocclusion proceeds in two steps. First, uncovered regions are estimated from the displaced frame difference. These uncovered regions are classified into actual disocclusions and false alarms by observing the motion characteristics of uncovered regions. Occlusion and disocclusion are considered as dual events and this relationship is explained in detail. The algorithm for detecting occlusion is developed by modifying the disocclusion detection algorithm in accordance with the duality principle. The overall tracking algorithm is computationally superior to existing region-based methods for object tracking. The immediate applications of the proposed tracking algorithm are video compression using MPEG-4 and content retrieval based on standards like H.264. Preliminary simulation results demonstrate the performance of the proposed algorithm.     

Adaptive visual tracking using the prioritized Q-learning algorithm: MDP-based parameter learning approach

This paper introduces an adaptive visual tracking method that combines the adaptive appearance model and the optimization capability of the Markov decision process. Most tracking algorithms are limited due to variations in object appearance from changes in illumination, viewing angle, object scale, and object shape. This paper is motivated by the fact that tracking performance degradation is caused not only by changes in object appearance but also by the inflexible controls of tracker parameters. To the best of our knowledge, optimization of tracker parameters has not been thoroughly investigated, even though it critically influences tracking performance. The challenge is to equip an adaptive tracking algorithm with an optimization capability for a more flexible and robust appearance model. In this paper, the Markov decision process, which has been applied successfully in many dynamic systems, is employed to optimize an adaptive appearance model-based tracking algorithm. The adaptive visual tracking is formulated as a Markov decision process based dynamic parameter optimization problem with uncertain and incomplete information. The high computation requirements of the Markov decision process formulation are solved by the proposed prioritized Q-learning approach. We carried out extensive experiments using realistic video sets, and achieved very encouraging and competitive results.     

视觉跟踪算法综述

随着信息技术与智能科学的迅速发展,计算机视觉已经成为IT产业和高新技术领域的前沿.视觉跟踪是当前计算机视觉领域的热点问题之一.阐述了视觉跟踪算法的研究现状,包括视觉跟踪算法的种类,常用数学方法,研究了基于区域的跟踪算法、基于模型的跟踪算法、基于特征的跟踪算法、基于主动轮廓的跟踪算法、参数估计方法和无参密度估计方法,并探讨了视觉跟踪算法的未来研究方向.     

Biologically adaptive robust mean shift algorithm with Cauchy predator-prey BBO and space variant resolution for unmanned helicopter formation

Visual tracking technology can provide measurement information for unmanned helicopter formation and thus, more attention is being paid to this research area. We propose a novel mean shift （MS） algorithm that is both adaptive and robust for unmanned helicopter formation and apply it to the leading unmanned helicopter tracking. The movement of an unmanned helicopter is very flexible and changeable, which makes the tracking there of more difficulty than for common targets. In creating an algorithm that can adapt to the acceleration of the unmanned helicopter and estimates both the scale and orientation of the movement changes, we combine the traditional MS with the bio-inspired Cauchy predator-prey biogeography-based optimization （CPPBBO） evolutionary algorithm, and also the space variant resolution （SVR） mechanism of the human visual system （MS-CPPBBO-SVR）. To demonstrate the effectiveness and robustness of the proposed method and justify the importance of the CPPBBO algorithm and SVR mechanism at the same time, a series of comparative exper- iments were carried out. The experimental results of the proposed MS-CPPBBO-SVR method are compared with other competitive tracking methods, such as MS, MS with SVR （MS-SVR）, MS-SVR with several other optimization algorithms, and the robust particle filter algorithm. The experimental results demonstrate that our proposed tracking approach, MS-CPPBBO-SVR, is more adaptive, robust and efficient in target tracking than the other methods.     

采用弱预测机制的人体运动跟踪算法

快速运动和自遮挡是人体运动跟踪的难点所在 .为此提出了一种采用弱预测机制的人体运动跟踪算法 .该算法首先通过全局搜索 ,确定候选人体特征集 ;然后建立特征的色彩、运动等属性的时变模型 ,构造贝叶斯分类器 ,实现特征对应 ;最后根据人体特征层次模型 ,检验特征匹配 ,并实现被遮挡特征的定位 .为提高跟踪效率 ,采用了基于图象多分辨率表示的特征搜索算法 ,由低分辨率图象通过全局搜索来获取初始候选特征集 ,然后在高分辨率下 ,不断改善候选特征精度 .实验结果表明 ,该算法能实现对快速人体运动的跟踪并有效解决自遮挡问题 .     

HIRBIR: A hierarchical approach to region-based image retrieval

This paper proposes a hierarchical approach to region-based image retrieval (HIRBIR) based on wavelet transform whose decomposition property is similar to human visual processing. First, automated image segmentation is performed fast in the low-low (LL) frequency subband of the wavelet domain that shows the desirable low image resolution. In the proposed system, boundaries between segmented regions are deleted to improve the robustness of region-based image retrieval against segmentation-related uncertainty. Second, a region feature vector is hierarchically represented by information in all wavelet subbands, and each feature component of a feature vector is a unified color–texture feature. Such a feature vector captures well the distinctive features (e.g., semantic texture) inside one region. Finally, employing a hierarchical feature vector, the weighted distance function for region matching is tuned meaningfully and easily, and a progressive stepwise indexing mechanism with relevance feedback is performed naturally and effectively in our system. Through experimental results and comparison with other methods, the proposed HIRBIR shows a good tradeoff between retrieval effectiveness and efficiency as well as easy implementation for region-based image retrieval.