交通信息采集系统中的行人检测算法

根据高速公路行人运动的先验知识,设计了一种基于视频检测技术的高速公路行人检测算法.该算法采用背景帧差分法获取运动目标区域,采用跟踪链实现运动目标跟踪,根据行人运动的先验知识在运动目标中检测行人.算法已嵌入到交通信息采集系统中,在高速公路上进行的现场测试结果表明,算法具有较好的实时性和实效性.     

基于Adaboost和码本模型的手扶电梯出入口视频监控方法

针对传统视频监控方法无法对密集前景目标进行准确分割的问题,提出一种基于Adaboost和码本模型的多目标视频监控方法。首先,通过训练得到Adaboost人头分类器,利用码本算法为垂直拍摄的手扶电梯出入口图像建立背景模型,提取前景图像对其进行人头检测和跟踪;之后,剔除行人目标得到物件目标,对物件目标进行跟踪;最后,根据行人和物件的运动特征进行监控。对12段出入口视频序列的实验结果表明,监控方法能够准确稳定地跟踪行人和物件,完成逆行检测、客流统计、行人拥堵和物件滞留等监控任务,处理速度达到36帧/秒,目标跟踪准确率达到94%以上,行为监控准确率达到95.8%,满足智能视频监控系统鲁棒性、实时性和准确性的要求。     

利用稀疏协同模型的目标跟踪算法

针对增强视频目标跟踪鲁棒性难题,提出一种利用稀疏协同判别模型和生成模型的跟踪算法.在判别模型中,利用先验视觉知识训练一个基于SIFT特征的过完备字典,用于构建目标外观模型和训练分类器实现目标与背景的分离;在生成模型中,提取目标的局部特征以及计算目标的遮挡信息来构建目标模板,通过计算候选目标与目标模板的相似度实现对目标的跟踪;最终利用乘性策略融合2种模型的跟踪结果.定性和定量的实验结果表明,与经典跟踪算法相比,该算法具有较好的鲁棒性.     

基于时间序列模型的粒子滤波行人跟踪算法研究

针对行人运动的随机性导致运动状态模型适应性差和人在行走过程中可能发生短时全部或局部遮挡导致行人跟踪算法精度较低的问题,提出基于时间序列模型的粒子滤波行人跟踪算法;建立了行人运动时间序列模型;给出了基于对视频序列初始帧的检测,确定行人的位置、宽高等作为跟踪先验信息的方法;由先验信息计算加权颜色直方图构建初始粒子群分布,并利用时间序列运动模型预测粒子在下一时刻的状态分布,并更新粒子权值;根据有效粒子的个数判断是否进行重采样;最后由所有粒子的加权和估计行人的运动状态;仿真实验表明:文中提出根据行人的运动轨迹时间序列运动模型可使行人的状态估计更准确,预测误差进一步减小,预测精度得到了提高.     

基于时空上下文的手势跟踪与识别

针对复杂背景下手势运动过程中出现的手势形态变化、遮挡、光照变化等问题,提出了一种基于时空上下文的手势跟踪与识别方法。使用机器学习方法离线训练手势样本分类器,实现对手势的检测和定位;利用时空上下文跟踪算法对动态手势进行跟踪,同时为了避免跟踪过程中出现的漂移、目标丢失等情况,使用手势检测算法对手势位置信息进行实时校准;根据手势运动轨迹对手势运动进行跟踪与识别。实验表明,提出的方法可以实现对手势运动快速、准确、连续识别,满足人机交互的要求。     

基于多特征子空间与核学习的行人再识别

行人再识别指的是在无重叠视域多摄像机监控系统中, 匹配不同摄像机视域中的行人目标.针对当前基于距离测度学习的行人再识别算法中存在着特征提取复杂、训练过程复杂和识别效果差的问题, 我们提出一种基于多特征子空间与核学习的行人再识别算法.该算法首先在不同特征子空间中基于核学习的方法得到不同特征子空间中的测度矩阵以及相应的相似度函数, 然后通过比较不同特征子空间中的相似度之和来对行人进行识别.实验结果表明, 本文提出的算法具有较高的识别率, 其中在VIPeR数据集上, RANK1达到了40.7%, 且对光照变化、行人姿态变化、视角变化和遮挡都具有很好的鲁棒性.     

MB-LBP特征提取和粒子滤波相结合的运动目标检测与跟踪算法研究

在复杂环境下,由于行人密度大以及运动随机性,导致运动目标(行人)难以检测和跟踪,造成人员计数误差。提出一种MB-LBP(Multi-scale Block Local Binary Pattern)特征提取和粒子滤波相结合的运动目标检测与跟踪算法来解决此问题。该算法首先用AdaBoost提取MB-LBP特征训练生成分类器进行人头检测,并根据人头目标尺寸变化范围去除部分误检,然后用改进的粒子滤波算法预测跟踪多个运动目标,最后对跟踪的运动目标进行计数。实验结果表明,提出的算法能够对复杂环境下多个运动目标进行有效检测及跟踪,准确、快速地对视频帧中的人员进行计数。     

相机运动条件下的行人检测与跟踪

行人检测与跟踪在司机辅助安全系统和视频监控等领域具有重要的地位.针对目前存在的关键问题,如人体运动,相机运动,背景及形状、角度等变化对检测及跟踪带来的干扰,提出了一种将运动信息与形状信息相结合的行人检测方法,准确检测运动摄像机拍摄的直立运动人体;使用了基于小面积目标的跟踪算法进行人体跟踪;利用实际拍摄的视频序列进行算法验证.实验结果表明,混合检测算法速度快,准确率高;基于小面积的跟踪算法能够鲁棒的跟踪检测到的运动人体.     

基于特征融合与分类器在线学习的目标跟踪算法

为了解决目标在复杂环境下表观变化引起的跟踪漂移问题,提出一种基于多特征融合与分类器在线学习的目标跟踪算法.该算法利用不同表观特征训练子分类器,通过构建损失函数求得各子分类器可信度,进而加权融合子预测结果,得到当前帧最佳目标状态估计;同时,依据最近-最远边界原则和协同训练理论粗更新训练样本集,并通过精选择准则得到更具代表性的训练样本集,实现子分类器自适应更新.实验结果表明,所提出的算法在多种典型测试场景中都能取得较鲁棒的跟踪效果.     

基于运动特征及位置估计的行人检测算法

根据行人运动的特点和行人在图像中位置与身高的对应关系,提出了一种结合运动特征与位置估计的行人检测算法。提取运动特征和聚合通道特征（ACF）,将提取的特征放到Real Adaboost分类器里进行训练,并对行人可能存在的位置建立评估模型;在检测阶段,通过分类器确定行人的候选区域,然后采用非极大值抑制算法去除重叠窗口,最后对行人候选区域应用位置评估模型进一步判断,以此排除可能的非行人目标。实验采用Caltech数据集对算法进行仿真,该算法的平均对数漏检率为27.12%,与ACF算法的相比降低了5.9个百分点。实验表明运动特征在视频检测中能够与静态特征进行信息互补,同时行人的位置估计在摄像机固定的情况下,具有一定的判断能力。     

基于模糊理论的行人异常动作检测

为在智能监控系统中自动识别行人的异常动作,提出简化的人体关节模型图。根据行人躯干和四肢轮廓角度的变化,设计用于模糊化的函数式。提出利用躯干和四肢的模糊隶属度通过计算来得到整个人异常度的一种基于模糊理论异常行为判别的算法。在系统实现中,提出利用质心轨迹和模糊判别的联合方法来甄别行人是否异常的方法。模糊判别可实现在视频监控范围内对行人行为的主动分析,从而能够对行人异常的动作做出识别并进行报警处理。通过实验证明该方法具有较高的识别率。     

基于改进的稀疏重构算法的行人异常行为分析

针对视频监控中行人异常行为识别问题,首先对行人进行跟踪,然后对跟踪得到的轨迹进行分析,最后判断行人行为是否存在异常。在行人跟踪方面,在时空上下文跟踪算法的基础上结合卡尔曼滤波器,有效改进了复杂背景中的遮挡问题。在异常分析方面,将跟踪得到的目标轨迹按照轨迹形状进行分类,得到场景中的正常轨迹集;将这些轨迹集作为后续处理的训练样本集,通过改进的稀疏重构算法对轨迹进行分析,利用重构误差来判断异常。五段视频序列的测试结果表明,该方法与改进前的方法相比具有较高的识别率。     

基于多尺度建议框的目标跟踪误差修正方法

随着深度学习技术引入视觉目标跟踪领域,目标跟踪算法的精度和鲁棒性有了很大的提高。但在低空无人机跟踪目标的实际场景中,情况比较复杂,如相机的抖动、大量的遮挡、视角和焦距的改变等,使得跟踪算法的准确性受到极大挑战。目前的算法大多建立在目标外观变化缓慢的前提假设下,在跟踪的过程中不具备检测和修复漂移（跟踪误差）的能力。针对该问题,提出了一种基于多尺度建议框的目标跟踪误差修正方法。离线阶段,利用大量的已标注的目标样本训练基于多尺度建议框的目标跟踪修正模型,获取不同类别目标的先验知识。在线阶段在核相关滤波跟踪的基础上,依据相关响应置信度自适应评价的结果,通过目标跟踪修正模型不定期重新初始化目标的位置,避免了因为误差累积而导致跟踪失败。算法在无人机航拍数据集上进行了测试,结果表明,该跟踪算法在目标发生较大形变的情况下能较好的修正跟踪漂移问题。相比于其他几种算法,目标跟踪的成功率和精度分别提高了14.3%和3.1%。     

基于改进DSST的行人遮挡跟踪算法

为解决判别尺度空间跟踪(DSST)算法在行人处于长期完全遮挡后又重新出现的情况下无法跟踪的问题, 提出了一种改进的跟踪算法(DDSST). 在DSST框架下,首先对行人目标跟踪, 然后, 引入高置信度指标计算策略作为跟踪准确可信度反馈机制, 在跟踪丢失时利用可变部件模型(DPM)对跟踪目标重新定位. 最后, 通过评估在线目标跟踪基准(OTB)数据集和实际环境拍摄的视频序列对DDSST算法准确性进行验证, 并与其他跟踪算法进行比较. 实验分析表明, 改进算法相较DSST的距离精度与成功率提高了4.1% 和6%, 相比其他算法性能更优, 且在形变、遮挡、平面外旋转、运动模糊和尺度变换等条件下跟踪更稳定.     

Recognition of pedestrian trajectories and attributes with computer vision and deep learning techniques

Analyzing the walking behavior of the public is vital for revealing the need for infrastructure design in a local neighborhood, supporting human-centric urban area development. Traditional walking behavior analysis practices relying on manual on-street surveys to collect pedestrian flow data are labor-intensive and tedious. On the contrary, automated video analytics using surveillance cameras based on computer vision and deep learning techniques appears more effective in generating pedestrian flow statistics. Nevertheless, most existing methods of pedestrian tracking and attribute recognition suffer from several challenging conditions, such as inter-person occlusion and appearance variations, which leads to ambiguous identities and hence inaccurate pedestrian flow statistics.Therefore, this paper proposes a more robust methodology of pedestrian tracking and attribute recognition, facilitating the analysis of pedestrian walking behavior. Specific limitations of a current state-of-the-art method are inferred, based on which several improvement strategies are proposed: 1) incorporating high-level pedestrian attributes to enhance pedestrian tracking, 2) a similarity measure integrating multiple cues for identity matching, and 3) a probation mechanism for more robust identity matching. From our evaluation using two public benchmark datasets, the developed strategies notably enhance the robustness of pedestrian tracking against the challenging conditions mentioned above. Subsequently, the outputs of trajectories and attributes are aggregated into fine-grained pedestrian flow statistics among different pedestrian groups. Overall, our developed framework can support a more comprehensive and reliable decision-making for human-centric planning and design in different urban areas. The framework is also applicable to exploiting pedestrian movement patterns in different scenes for analyses such as urban walkability evaluation. Moreover, the developed mechanisms are generalizable to future researches as a baseline, which provides generic insights of how to fundamentally enhance pedestrian tracking.     

基于混合跟踪模型的室内步行人体3D运动估计

针对步行人体3D运动估计过程中的自遮挡问题, 提出了基于混合跟踪模型的粒子滤波算法. 首先, 利用自遮挡状态检测模型, 将步行人体运动划分为四种自遮挡状态; 其次, 根据混合跟踪模型, 针对不同的自遮挡状态, 算法采用不同的跟踪模型; 最后, 为了估计遮挡状态下的人体运动, 算法提出了基于M-估计的在线训练方法 以训练肢体运动相关系数. 经过实验分析, 算法对处 于自遮挡状态下的人体3D运动估计有着良好的效果, 人体3D运动的估计精度得到了提高.     

基于道路环境上下文的行人跟踪方法

针对目前城市交通中人车混行场景中行人跟踪效果不佳的问题,提出了一种基于道路环境上下文的行人跟踪方法。首先通过对道路环境上下文进行分析,建立道路模型;其次在道路模型的约束下建立行人与环境的交互运动模型;最后利用该模型进行行人的跟踪。在真实场景中的实验表明使用了道路上下文信息的跟踪方法与连续离散连续能量最小化的多行人跟踪方法相比,多目标跟踪准确度从47.6%提升至63.2%,多目标跟踪精度从68.8%提升至74.3%。数值结果表明道路上下文信息对于提高人车混行场景中行人跟踪效果的有效性。     

Tracking pedestrians using local spatio-temporal motion patterns in extremely crowded scenes

Tracking pedestrians is a vital component of many computer vision applications, including surveillance, scene understanding, and behavior analysis. Videos of crowded scenes present significant challenges to tracking due to the large number of pedestrians and the frequent partial occlusions that they produce. The movement of each pedestrian, however, contributes to the overall crowd motion (i.e., the collective motions of the scene's constituents over the entire video) that exhibits an underlying spatially and temporally varying structured pattern. In this paper, we present a novel Bayesian framework for tracking pedestrians in videos of crowded scenes using a space-time model of the crowd motion. We represent the crowd motion with a collection of hidden Markov models trained on local spatio-temporal motion patterns, i.e., the motion patterns exhibited by pedestrians as they move through local space-time regions of the video. Using this unique representation, we predict the next local spatio-temporal motion pattern a tracked pedestrian will exhibit based on the observed frames of the video. We then use this prediction as a prior for tracking the movement of an individual in videos of extremely crowded scenes. We show that our approach of leveraging the crowd motion enables tracking in videos of complex scenes that present unique difficulty to other approaches.     

An adaptable neural-network model for recursive nonlinear traffic prediction and modeling of MPEG video sources

Multimedia services and especially digital video is expected to be the major traffic component transmitted over communication networks [such as internet protocol (IP)-based networks]. For this reason, traffic characterization and modeling of such services are required for an efficient network operation. The generated models can be used as traffic rate predictors, during the network operation phase (online traffic modeling), or as video generators for estimating the network resources, during the network design phase (offline traffic modeling). In this paper, an adaptable neural-network architecture is proposed covering both cases. The scheme is based on an efficient recursive weight estimation algorithm, which adapts the network response to current conditions. In particular, the algorithm updates the network weights so that 1) the network output, after the adaptation, is approximately equal to current bit rates (current traffic statistics) and 2) a minimal degradation over the obtained network knowledge is provided. It can be shown that the proposed adaptable neural-network architecture simulates a recursive nonlinear autoregressive model (RNAR) similar to the notation used in the linear case. The algorithm presents low computational complexity and high efficiency in tracking traffic rates in contrast to conventional retraining schemes. Furthermore, for the problem of offline traffic modeling, a novel correlation mechanism is proposed for capturing the burstness of the actual MPEG video traffic. The performance of the model is evaluated using several real-life MPEG coded video sources of long duration and compared with other linear/nonlinear techniques used for both cases. The results indicate that the proposed adaptable neural-network architecture presents better performance than other examined techniques.     

Modeling local behavior for predicting social interactions towards human tracking

Human interaction dynamics are known to play an important role in the development of robust pedestrian trackers that are needed for a variety of applications in video surveillance. Traditional approaches to pedestrian tracking assume that each pedestrian walks independently and the tracker predicts the location based on an underlying motion model, such as a constant velocity or autoregressive model. Recent approaches have begun to leverage interaction, especially by modeling the repulsion forces among pedestrians to improve motion predictions. However, human interaction is more complex and is influenced by multiple social effects. This motivates the use of a more complex human interaction model for pedestrian tracking. In this paper, we propose a novel human tracking method by modeling complex social interactions. We present an algorithm that decomposes social interactions into multiple potential interaction modes. We integrate these multiple social interaction modes into an interactive Markov Chain Monte Carlo tracker and demonstrate how the developed method translates into a more informed motion prediction, resulting in robust tracking performance. We test our method on videos from unconstrained outdoor environments and evaluate it against common multi-object trackers.