基于Kalman和双级联随机森林的在线目标跟踪算法

针对传统的基于检测的在线目标跟踪算法容易产生跟踪漂移的现象,提出了一种新的在线目标跟踪算法.以基于主方向模板特征的双级联随机森林分类器作为检测器,卡尔曼滤波器作为跟踪器.首先利用卡拉曼算法跟踪目标,然后以跟踪的目标位置为中心向外扩展一定的范围作为双级联随机森林分类器的检测区域,利用全局随机森林分类器和局部随机森林分类器进行目标检测,并将检测结果作为Kalman跟踪算法下一帧的观测值.实验结果显示,提出的算法在跟踪大小420×320的图像时,跟踪速度达到24.3 f/s(帧/秒),目标中心位置误差在30 pixel时,算法准确率可达到80％以上.     

基于增强群跟踪器和深度学习的目标跟踪

为解决基于外观模型和传统机器学习目标跟踪易出现目标漂移甚至跟踪失败的问题,该文提出以跟踪-学习-检测(TLD)算法为框架,基于增强群跟踪器(FoT)和深度学习的目标跟踪算法。FoT实现目标的预测与跟踪,增添基于时空上下文级联预测器提高预测局部跟踪器的成功率,快速随机采样一致性算法评估全局运动模型,提高目标跟踪的精确度。深度去噪自编码器和支持向量机分类器构建深度检测器,结合全局多尺度扫描窗口搜索策略检测可能的目标。加权P-N学习对样本加权处理,提高分类器的分类精确度。与其它跟踪算法相比较,在复杂环境下,不同图片序列实验结果表明,该算法在遮挡、相似背景等条件下具有更高的准确度和鲁棒性。     

动态背景下运动车辆的检测与跟踪研究

为实现对车载设备视频图像中车辆的识别和跟踪,针对图像中的运动目标和动态背景,提出了一种基于特征学习的目标检测和超像素跟踪算法.该算法首先对训练图像进行HOG特征提取,并利用AdaBoost算法得到强分类器.利用强分类器对采集的图像进行车辆检测,从而确定搜索区域.结合对搜索区域的超像素分割结果,采用均值漂移聚类算法实现车辆识别与跟踪.实验结果表明,该算法可以很好地实现视频序列中的车辆识别,提高了目标跟踪的实时性.     

融合GMS检测和置信度判别的TLD目标跟踪

针对目标在遮挡、尺度变化等复杂场景下易产生模型漂移问题,基于跟踪学习检测(TLD)框架提出一种结合基于网格的运动统计(GMS)检测和置信度判别的长时目标跟踪算法。首先在跟踪模块中采用快速判别尺度空间的相关滤波器(fDSST)作为跟踪器,利用位置滤波器和尺度滤波器对上一帧目标进行位置与尺度的判别,并依据TLD算法中跟踪模块与检测模块的独立性,将跟踪模块结果输入检测模块中,采用平均峰值相关能量(APCE)对模板更新进行置信度判别。在检测模块中先引入GMS网格运动统计作为检测器,使具有快速旋转不变性特征的ORB(OrientedFASTandRotatedBRIEF)算法对上一帧目标进行特征匹配,再利用网格运动统计对匹配结果进行过滤,实现目标位置的粗定位,依据预测位置对目标检测区域进行适当的动态缩减,最后使用级联分类器对目标进行精准定位。结果表明,本文提出的跟踪方法在有效防止模型漂移的情况下,大大提高了算法的跟踪速度,同时对目标遮挡、尺度变化及旋转等挑战环境也具有较好的准确性和鲁棒性。     

融合粒子滤波和在线adaboost分类器的目标跟踪方法研究

传统的粒子滤波由于粒子数目有限,且模板会发生变化,模板更新时会出现漂移甚至导致跟踪失败。文章提出了融合粒子滤波和在线adaboost分类器的目标跟踪方法。将粒子滤波的跟踪结果作为正样本更新样本集,再通过样本集训练分类器检测到目标位置。然后将检测和跟踪中具有较高置信度的结果作为最终目标位置。实验证明,这种方式可以很好地解决目标重现和漂移问题。     

基于强化学习的公共空间行人轨迹跟踪

对公共空间中的多目标行人轨迹跟踪问题,提出一种基于强化学习的多目标行人轨迹跟踪算法。首先采用高精确度的目标检测器检测公共空间视频中的行人目标,并为每个目标分配一个独立的单目标跟踪器进行轨迹跟踪;将每个目标作为独立智能体,通过深度强化学习方式进行训练;接下来结合跟踪轨迹与检测目标之间的表观和位置特征构建相似度代价矩阵;最终通过匈牙利算法实现数据关联。实验表明,在常用公开数据集上本文算法跟踪精确度达76.1%,表明算法对多目标轨迹跟踪的可行性与有效性。     

基于改进TLD算法的激光视觉传感型焊缝跟踪

为了解决基于线激光视觉传感的焊缝中心位置定位精度不高的问题, 采用了一种基于改进跟踪-学习-检测(TLD)算法的焊缝跟踪方法。由激光视觉传感器实时获取焊缝图像, 采用将跟踪器与检测器结合的TLD算法实时跟踪焊缝特征点, 同时通过在线学习机制更新分类器参量。在此基础上对激光条纹图像截取感兴趣区域, 大幅减少检测器的搜索区域; 根据激光条纹光强分布特性, 结合纠偏方向选取跟踪器有效特征点, 以此提高算法效率, 对不锈钢板V型焊缝和搭接焊缝进行跟踪试验。结果表明, 跟踪与检测可实现共同定位焊缝中心位置, 其融合的焊缝跟踪方法能够准确地提取焊缝特征点, 两种焊缝跟踪平均绝对误差分别为0.062mm和0.052mm。此方法为提高焊缝跟踪精度提供了依据。     

LK光流法的多信息融合目标跟踪算法研究

针对改善动态目标物体跟踪过程中特征点选取的准确性及实时性问题,提出一种基于LK光流法的多信息融合目标跟踪方法并设计相关算法。将Sift提取出的特征点作为初始特征点,结合Harris角点检测函数R值,通过设定阈值,从中选取出具有高对比度的特征点作为目标特征点,对其进行光流法跟踪;然后引入在线学习检测器,使得在跟踪器跟踪失败之后可以更新目标位置,以实现稳定、准确跟踪。实验的结果从动态目标识别所需时间、特征点的数量、交并比以及每秒处理的帧数几个方面进行比较分析,证明所提算法能够实现准确跟踪,且实时性强。     

基于在线半监督boosting的协同训练目标跟踪算法

基于自训练的判别式目标跟踪算法使用分类器的预测结果更新分类器自身,容易累积分类错误,从而导致漂移问题。为了克服自训练跟踪算法的不足,该文提出一种基于在线半监督boosting的协同训练目标跟踪算法(简称Co-SemiBoost),其采用一种新的在线协同训练框架,利用未标记样本协同训练两个特征视图中的分类器,同时结合先验模型和在线分类器迭代预测未标记样本的类标记和权重。该算法能够有效提高分类器的判别能力,鲁棒地处理遮挡、光照变化等问题,从而较好地适应目标外观的变化。在若干个视频序列的实验结果表明,该算法具有良好的跟踪性能。     

基于Kalman滤波与Camshift算法的水面目标跟踪

根据水面监控图像的特点,对运动载体采集到的水面视频图像进行处理,从而实现对运动目标的跟踪。首先,利用Haar分类器检测出水面的运动目标,并用检测结果初始化Camshift跟踪器的搜索窗口;然后,运用Kalman滤波器与Camshift组合算法实现对运动目标的跟踪。其中,利用Kalman滤波算法预测目标在下一帧中出现的位置,Camshift算法用来跟踪目标,以此减小搜索范围,提高跟踪效率。实验结果表明,该算法能够实现对水面运动舰船的检测并进行有效跟踪。     

Visual object tracking with online sample selection via lasso regularization

In the past years, discriminative methods are popular in visual tracking. The main idea of the discriminative method is to learn a classifier to distinguish the target from the background. The key step is the update of the classifier. Usually, the tracked results are chosen as the positive samples to update the classifier, which results in the failure of the updating of the classifier when the tracked results are not accurate. After that the tracker will drift away from the target. Additionally, a large number of training samples would hinder the online updating of the classifier without an appropriate sample selection strategy. To address the drift problem, we propose a score function to predict the optimal candidate directly instead of learning a classifier. Furthermore, to solve the problem of a large number of training samples, we design a sparsity-constrained sample selection strategy to choose some representative support samples from the large number of training samples on the updating stage. To evaluate the effectiveness and robustness of the proposed method, we implement experiments on the object tracking benchmark and 12 challenging sequences. The experiment results demonstrate that our approach achieves promising performance.     

Real‐Time Apartment Building Detection and Tracking with AdaBoost Procedure and Motion‐Adjusted Tracker

In this letter, we propose a novel approach to detecting and tracking apartment buildings for the development of a video‐based navigation system that provides augmented reality representation of guidance information on live video sequences. For this, we propose a building detector and tracker. The detector is based on the AdaBoost classifier followed by hierarchical clustering. The classifier uses modified Haar‐like features as the primitives. The tracker is a motion‐adjusted tracker based on pyramid implementation of the Lukas‐Kanade tracker, which periodically confirms and consistently adjusts the tracking region. Experiments show that the proposed approach yields robust and reliable results and is far superior to conventional approaches.     

Integration of Bayes detection with target tracking

Existing detection systems generally are operated using a fixed threshold and optimized to the Neyman-Pearson criterion. An alternative is Bayes detection, in which the threshold varies according to the ratio of prior probabilities. In a recursive target tracker such as the probabilistic data association filter (PDAF), such priors are available in the form of a predicted location and associated covariance; however, the information is not at present made available to the detector. Put another way, in a standard detection/tracking implementation, information flows only one way: from detector to tracker. Here, we explore the idea of two-way information flow, in which the tracker instructs the detector where to look for a target, and the detector returns what it has found, more specifically, we show that the Bayesian detection threshold is lowered in the vicinity of the predicted measurement, and we explain the appropriate modification to the PDAF. The implementation is simple, and the performance is remarkably good     

Robust object tracking via deformation samples generator

In object tracking applications, it is common for trackers to experience drift problems when the object of interest becomes deformed, which compromises the ability of the tracker to track the object. It is therefore desirable to develop a learning tracker classifier that is robust to deformations. The performance of existing trackers that employ deep classification networks degrades when the amount of training data is limited and does not cover all possible scenarios. While these limitations can be mitigated in part by using larger training datasets, these datasets may still not cover all situations and the positive samples are still monotonous. To overcome this problem, we propose a novel deformation samples generator that generates samples that would normally be difficult for the tracker to classify. In the proposed framework, both the classifier and deformation samples generator learn in a joint manner. Our experiments show that the proposed approach outperforms state-of-the-art methods in both quantitative and qualitative evaluations for the visual object tracking task.     

基于多实例回归模型的视觉跟踪算法研究

目前大部分基于检测的跟踪算法将跟踪任务看作是一个类别分类的任务,当目标发生形变或者遇到相似物体的干扰时,容易导致模型漂移。为此该文提出一种多实例回归跟踪算法。在该算法中,跟踪任务被认为建立在实例模型之上更为合适,为此该文利用一帧图像建立实例模型,并在时间序列上建立多实例模型集合表征目标的最近状态;为使跟踪算法能够适应目标的形变,利用逻辑回归将实例模型作为隐变量,由最近若干帧建立的正负样本集作为训练集,共同构建多实例回归跟踪模型。由于跟踪模型在整体上对多个实例模型建模,把它们紧密地联系在一起,故能有效应对目标的形变;由于模型漂移仅会影响当前帧的实例模型,各个实例模型之间互相独立,故跟踪算法能够有效减轻模型漂移对鲁棒跟踪的影响。实验中,OTB 2013数据库和UAV 123数据库被用来验证该文算法,DeepSRDCF, Siamese-fc等算法作为对比算法,实验结果表明,该文算法不仅充分发挥了基于多实例回归模型进行跟踪的优势,在形变等属性上具有很好的性能,而且在整体性能上优于各类先进算法3%~5%。     

基于工地场景的深度学习目标跟踪算法

针对施工现场环境复杂,难以高效管理的问题.提出了基于工地场景的深度学习目标跟踪算法,辅助施工顺利进行.根据工地现场目标的连续性,构建增强群跟踪器,提升目标成功跟踪的概率.然后从滑动窗口、Stacked Denoising Auto Encoder（SDAE）和Support Vector Machine（SVM）三方面组建深度检测器.在滑动窗口方面：从梯度角度建立模型实现窗口自适应.在SDAE算法方面：构建反向算法微调网络参数.优化SVM算法降低跟踪时目标漂移和跟踪失败的概率,最终实现目标高精度跟踪.通过实验表明本文提出的算法可有效对目标进行跟踪,实现动态管理.     

基于相关滤波的红外目标跟踪抗遮挡处理

针对传统的相关滤波算法在红外目标跟踪过程中,目标被完全遮挡后跟踪失效的问题,提出一种结合了多尺度滤波跟踪器和基于深度学习检测器的目标实时跟踪抗遮挡算法.首先使用跟踪器跟踪目标,计算目标的峰值响应强度并比较峰值响应强度与经验阈值的大小以判断目标是否被遮挡或跟踪丢失.然后当目标被遮挡或跟踪丢失时,停止更新跟踪器,由于目标被...     

A NEW MODELING AND FILTERING APPROACH FOR TRACKING MANEUVERING TARGETS

A new modeling and filtering approach for tracking maneuvering targets is presented in thispaper.The approach,which makes optimal estimate for the model With the random variable possible,depends on random step modeling of target maneuvers.In the new model,the unknown targetacceleration is treated as a random variable and then estimated directly.A detector is designed tofind out the target maneuvers and the estimation algorithm will be restarted when the maneuvers oc-cur.Combination of three-dimention Kalman filter with a detector forms a tracker for maneuveringtargets.The new tracking scheme is easy to implement and its capability is illustrated in two trackingexamples in which the new approach is compared with Mooses'on the performance.     

Autonomous mobile target tracking system based on grey-fuzzycontrol algorithm

This paper presents a new position-based tracking system for autonomous mobile target tracking task. A grey-fuzzy controller (GFC) is developed for motion control of the tracker, in which dynamics models of the target and tracker are not required a priori. The target detection is based on the adaptive visual detector (AVD), which can online adjust the histogram model based on the change of surrounding conditions, such as light variation, in a natural environment. The AVD and GFC are integrated together for mobile target-tracking applications. There are several advantages of the integrated system, in particular: (1) it can rapidly learn the target appearance model for the detection involved with the tracking task; (2) the temporal dynamics model of the target motion can be approximated for the predictive localization of the moving target; and (3) the system can deal with the uncertain environmental conditions to ensure the tracking performance by GFC. Three mobile robots in the authors' laboratory have been used to demonstrate the success of this integrated system experimentally. They also conduct target tracking experiments, in which Chung Cheng-I tracks various moving targets. The results demonstrate the robustness and flexibility of the overall system in dealing with mobile target-tracking problems under varied natural environment conditions