Probabilistic Modeling of Scene Dynamics for Applications in Visual Surveillance

We propose a novel method to model and learn the scene activity, observed by a static camera. The proposed model is very general and can be applied for solution of a variety of problems. The motion patterns of objects in the scene are modeled in the form of a multivariate nonparametric probability density function of spatiotemporal variables (object locations and transition times between them). Kernel Density Estimation is used to learn this model in a completely unsupervised fashion. Learning is accomplished by observing the trajectories of objects by a static camera over extended periods of time. It encodes the probabilistic nature of the behavior of moving objects in the scene and is useful for activity analysis applications, such as persistent tracking and anomalous motion detection. In addition, the model also captures salient scene features, such as the areas of occlusion and most likely paths. Once the model is learned, we use a unified Markov Chain Monte Carlo (MCMC)-based framework for generating the most likely paths in the scene, improving foreground detection, persistent labeling of objects during tracking, and deciding whether a given trajectory represents an anomaly to the observed motion patterns. Experiments with real-world videos are reported which validate the proposed approach.     

Robust foreground detection in video using pixel layers

A framework for robust foreground detection that works under difficult conditions such as dynamic background and moderately moving camera is presented in this paper. The proposed method includes two main components: coarse scene representation as the union of pixel layers, and foreground detection in video by propagating these layers using a maximum-likelihood assignment. We first cluster into "layers" those pixels that share similar statistics. The entire scene is then modeled as the union of such non-parametric layer-models. An in-coming pixel is detected as foreground if it does not adhere to these adaptive models of the background. A principled way of computing thresholds is used to achieve robust detection performance with a pre-specified number of false alarms. Correlation between pixels in the spatial vicinity is exploited to deal with camera motion without precise registration or optical flow. The proposed technique adapts to changes in the scene, and allows to automatically convert persistent foreground objects to background and re-convert them to foreground when they become interesting. This simple framework addresses the important problem of robust foreground and unusual region detection, at about 10 frames per second on a standard laptop computer. The presentation of the proposed approach is complemented by results on challenging real data and comparisons with other standard techniques.     

Unsupervised motion detection with background update and shadow suppression

An algorithm is developed to detect moving object and suppress shadow.According to motion variations caused by some moving objects in a scene,a background update approach is proposed.The developed update method efficiently prevents undesired corruption of background and does not consider the adaptation coefficient or the learning rate used in some existing algorithms.A multi-scale wavelet transform methodology is used to segment foreground from a clutter background.The optimal selection of threshold value is automatically determined which does not require any complex supervised training or manual calibration.According to photometric invariant,a color ratio difference is proposed to suppress shadow.Some complete foreground motion object regions are extracted by integrating moving object segmentation in the multi-scale wavelet with shadow suppression in the color ratio difference.The mentioned method is less affected by the presence of moving objects in a scene.Experimental results show that the proposed approach is efficient in detecting motion objects and suppressing shadows by comparisons.     

基于概率算法自适应更新背景的运动车辆检测

交通流量检测是智能交通系统中的一个重要研究方向和热点问题,基于视频的车辆检测是交通流量采集分析的核心技术,它为交通流量参数的实时获取提供了可能。为实现在复杂交通视频场景中实时准确检测各类的运动车辆,在研究传统背景差分算法的缺点的工作基础上,提出一个自适应的贝叶斯概率背景检测算法,进而完成了较准确的运动车辆分类检测。实验结果表明该方法具有高效实时的特点,能够较准确地实现复杂交通路面的背景提取和运动车辆的检测,具有良好的鲁棒性。     

Efficient region-based motion segmentation for a video monitoring system

This paper presents an efficient region-based motion segmentation method for segmentation of moving objects in a traffic scene with a focus on a video monitoring system (VMS). The presented method consists of two phases: first, in the motion detection phase, the positions of moving objects in a scene are determined using an adaptive thresholding method. To detect varying regions by moving objects, instead of determining the threshold value manually, we use an adaptive thresholding method to automatically choose the threshold value. Second, in the motion segmentation phase, pixels that have similar intensity and motion information are segmented using a weighted k-means clustering algorithm to the binary region of the motion mask obtained in the motion detection. In this way, we need not process a whole image so computation time is reduced. Experimental results demonstrate robustness not only in the variation of luminance conditions and changes in environmental conditions, but also for occlusions among multiple moving objects.     

交通场景中的实时多目标跟踪方法

针对多目标检测与跟踪技术的应用要求,提出一种用于交通场景的实时多目标跟踪方法。设计背景建模算法和基于角点动态特征的分层跟踪算法,利用背景建模算法提取视频帧前景,并在前景区域运用分层跟踪算法,包括Corner、Cluster和Object 3层架构,通过在不同层之间引入对应的聚类算法,以实现目标跟踪。实验结果表明,该方法适用于多数交通场景,对光照、阴影具有较强的鲁棒性。     

基于区域相关性的全局运动视频中非刚性对象跟踪

提出一种适合全局运动视频中自动探测与跟踪非刚性对象的OT-GAV模型.该模型首先利用基于区域相关性的RDM算法计算相邻帧区域匹配,并结合Q学习与K-S统计法优化匹配结果,获得较为精确的区域运动向量.然后,利用前景和背景存在的运动形态差异,区域动态纹理一致性及对象运动过程中保持区域完整性的特点,逐步实现前景对象区域的探测与合并.实验证明,本模型及其相关算法可在室内和室外环境下,自动探测前景关注对象,获得其较为精确的边缘信息,并实施有效的跟踪.同时,该模型还能够解决对象跟踪过程中的"空洞"问题.     

视频分析技术在日志系统中的应用

近年来,以多传感器信息融合为特征的日志系统成为了一个新的热点问题。日志系统使用者携带便携式摄像头所获得的视频信息是一类主要信息,获取其关键帧序列对完善系统功能具有重要意义。与其他应用相比,日志系统视频更加复杂多变。基于以上条件,本文提出非参数模型的方法估计背景运动;针对前景运动物体检测和跟踪问题,本文在非监督模式识别方法的基础上,使用统计、时空、颜色特征等给出前景区域判断;最后,利用上述结果得到关键帧序列。     

基于SAD与UKF-MeanShift的主动目标跟踪

针对复杂场景下动态目标难以准确分割以及目标难以准确定位的问题,提出将绝对差值和(SAD)方法、无迹卡尔曼滤波(UKF)和Mean shift算法相结合的混合自主跟踪动态目标的方法。首先,采用SAD方法获相邻两帧的视差信息,利用视差实现动态目标的检测,并依此建立目标的核直方图描述模型和状态空间模型,然后UKF算法对状态空间进行滤波估计,最后采用Mean shift 算法精确定位目标。实验结果表明该方法不仅能有效检测场景的动态目标,同时还能获得目标的运动信息。文中所提出的基于UKF-Mean shift的跟踪策略与相关算法相比,体现出较好的跟踪效果与时间性能。     

基于邻域相关性和帧间连续性的前景目标分割

提出了一种新的运动目标分割算法。首先利用像素的颜色、空间的和帧间的特性信息结合贝叶斯判别定理对视频图像进行粗分割，得到一个前景目标的二值图，由于该类方法基于像素间彼此独立的假设，导致分割出的前景目标不完整存在很多空洞。其次，基于前景目标局部邻域空间的一致性假设，计算该邻域内像素间的互相关系数；同时，基于背景的帧间连续性和前景的不连续性，计算像素帧间的互相关系数。最后，依据像素的互相关系数在该邻域内进行二次判决，以填补粗分割中前景目标内部的空洞。实验表明，在复杂背景交通视频中该分割算法具有较强的鲁棒性，并能获得更完整准确的前景目标。     

Motion-based skin region of interest detection with a real-time connected component labeling algorithm

This paper presents a motion-based skin Region of Interest (ROI) detection method using a real-time connected component labeling algorithm to provide real-time and adaptive skin ROI detection in video images. Skin pixel segmentation in video images is a pre-processing step for face and hand gesture recognition, and motion is a cue for detecting foreground objects. We define skin ROIs as pixels of skin-like color where motion takes place. In the skin color estimation phase, RGB color histograms are utilized to define the skin color distribution and specify the threshold to segment skin-like regions. A parallel computed connected component labeling algorithm is also proposed to group the segmentation results into several clusters. If a cluster covers any motion pixel, this cluster is identified as a skin ROI. The method’s results for real images are shown, and its speed is evaluated for various parameters. This technology is compatible with monitoring systems, scene understanding, and natural user interfaces.     

子空间描述的关节式目标跟踪

针对关节式目标变化对子空间描述造成的影响,本文提出了一种基于增量学习的关节式目标跟踪算法.该算法通过引入图像分割方法与快速傅里叶变换可有效消除背景像素对目标描述造成的影响以及目标区域前景目标位置对不准造成的误差,同时应用局部二值模式增加目标描述中像素点间的几何位置信息,应用基于增量学习的方法实现目标特征的在线更新,最终为跟踪算法提供较为精确的目标描述.实验结果表明,本文提出的关节式目标跟踪算法具有较好的目标跟踪效果.     

一种用于智能监控的目标检测和跟踪方法*

在对现有目标检测、跟踪算法进行分析对比的基础上,设计并实现了一种简单有效的目标检测和跟踪方案。首先提出了一种基于像素灰度归类和单模态高斯模型的背景重构算法,能够利用多帧包含前景目标的场景图像重构准确的背景模型。进而以此为基础采用背景减法进行各帧中目标的检测,并选取形心作为匹配特征实现了场景中多个目标的有效跟踪。实验表明,该方法实现简单,无须事先提供背景图像即可实现目标的准确检测和跟踪,其性能明显优于传统基于时间平均背景模型的方法。     

监控场景中的运动物体提取技术研究

针对监控场景中因存在遮挡而无法有效地提取出完整的运动序列这一问题,提出了一种将ViBe前景检测算法和改进后的粒子滤波跟踪算法相结合的跟踪提取方法。首先用ViBe来提取出场景中所有运动物体的前景轮廓;其次用粒子滤波来检测和跟踪目标物体;最后通过与目标物体的关联轮廓求交运算以及跟踪区域的反馈调节完成对目标物体运动帧序列的提取。当运动物体发生遮挡时,采用将跟踪区域内所检测到的前景轮廓重新加入到目标物体的关联轮廓中以保证后续可以继续用关联轮廓交集来提取。实验结果表明,该方法能够很好地保证提取的质量,并有效地解决了局部遮挡与全局遮挡情况下运动物体完整运动序列的提取。     

A network of sensor-based framework for automated visual surveillance

This paper presents an architecture for sensor-based, distributed, automated scene surveillance. The goal of the work is to employ wireless visual sensors, scattered in an area, for detection and tracking of objects of interest and their movements through application of agents. The architecture consists of several units known as Object Processing Units (OPUs) that are wirelessly connected in a cluster fashion. Cluster heads communicate with the Scene Processing Units which are responsible for analyzing all the information sent by the former. Object detection and tracking is performed by cooperative agents, named as Region and Object Agents. The area under surveillance is divided into several sub-areas. One camera is assigned to each sub-area. A Region Agent (RA) is responsible for monitoring a given sub-area. First, a background subtraction is performed on the scene taken by the camera. Then, a computed foreground mask is passed to the RA, which is responsible for creating Object Agents dedicated to tracking detected objects. Object detection and tracking is done automatically and is performed on the OPU. The tracking information and foreground mask are sent to a Scene Processing Unit that analyzes this information and determines if a threat pattern is present at the scene and performs appropriate action.     

基于背景减除法的视频序列运动目标检测

视频序列中运动目标的检测是目标识别、标记和追踪的重要组成部分,背景减除法是运动目标检测中被广泛应用的算法。针对光线变化、噪声和局部运动等影响运动目标检测效果的问题,提出一种基于背景减除法的视频序列运动目标检测算法。该算法结合背景减除法和帧间差分法,对当前帧像素点的运动状态进行判断,分别对静止和运动的像素点进行替换和更新,采用最大类间方差（Otsu）法对差分图像进行目标提取,并使用数学形态学运算去除目标中的噪声和冗余信息。实验结果表明,所提算法对于视频序列中运动目标的检测具有较好的视觉效果和较高的准确度,能够克服局部运动以及噪声等缺陷。     

目标检测算法在交通场景中应用综述

目标检测是计算机视觉领域的重要研究任务,在机器人、自动驾驶、工业检测等方面应用广泛。在深度学习理论的基础上,系统性总结了目标检测算法的发展与研究现状,对两类算法的特点、优缺点和实时性进行对比。以交通场景中三类典型物体（非机动车、机动车和行人）为目标,从传统检测方法、目标检测算法、目标检测算法优化、三维目标检测、多模态目标检测和重识别六个方面分别论述和总结目标检测算法检测识别交通场景目标的研究现状与应用情况,重点介绍了各类方法的优势、局限性和适用场景。归纳了常用目标检测和交通场景数据集及评价标准,比较分析两类算法性能,展望目标检测算法在交通场景中应用研究的发展趋势,为智能交通、自动驾驶提供研究思路。     

一种基于Kalman滤波的视频对象跟踪方法

为了更加准确地预测对象的位置和运动，利用刚体运动模型导出最佳Kalman系数，通过Kalman反馈滤波器对Moscheni等人提出的视频对象分割与跟踪算法进行改进，提出了一种将离散Kalman滤波技术用于视频序列的对象跟踪方法。这种方法可用于有关场景描述的各种应用领域中，如在机器视觉的研究中，对动态场景进行分析与理解；在基于对象的视频编码中（如MPEG-4），对视频对象进行分割后，分别进行编码，从而改善编码的可分级性及编码效率。实验结果表明，采用这种方法可以有效地改善时间-空间分割和目标跟踪，有助于更好地理解动态场景，并表现出良好的鲁棒性。     

基于大数据分析的移动对象轨迹预测方法

对移动对象的轨迹预测将在移动目标跟踪识别中具有较好的应用价值。移动对象轨迹预测的基础是移动目标运动参量的采集和估计,移动目标的运动参量信息特征规模较大,传统的单分量时间序列分析方法难以实现准确的参量估计和轨迹预测。提出一种基于大数据多传感信息融合跟踪的移动对象轨迹预测算法。首先进行移动目标对象进行轨迹跟踪的控制对象描述和约束参量分析,对轨迹预测的大规模运动参量信息进行信息融合和自正整定性控制,通过大数据分析方法实现对移动对象运动参量的准确估计和检测,由此指导移动对象轨迹的准确预测,提高预测精度。仿真结果表明,采用该算法进行移动对象的运动参量估计和轨迹预测的精度较高,自适应性能较强,稳健性较好,相关的指标性能优于传统方法。     

基于深度学习方法的复杂场景下车辆目标检测

近年来,深度学习算法逐渐尝试应用于目标检测领域。本文针对实际交通场景下的车辆目标,应用深度学习目标分类算法中具有代表性的Faster R-CNN框架,结合ImageNet中的车辆数据集,把场景中的目标检测问题转化为目标的二分类问题,进行车辆目标的检测识别。相比传统机器学习目标检测算法,基于深度学习的目标检测算法在检测准确度和执行效率上优势明显。通过本实验结果分析表明,该方法在识别精度以及速度上均取得了显著的提高。