SOINN-Based Abnormal Trajectory Detection for Efficient Video Condensation

With the evolution of video surveillance systems, the requirement of video storage grows rapidly; in addition, safe guards and forensic officers spend a great deal of time observing surveillance videos to find abnormal events. As most of the scene in the surveillance video are redundant and contains no information needs attention, we propose a video condensation method to summarize the abnormal events in the video by rearranging the moving trajectory and sort them by the degree of anomaly. Our goal is to improve the condensation rate to reduce more storage size, and increase the accuracy in abnormal detection. As the trajectory feature is the key to both goals, in this paper, a new method for feature extraction of moving object trajectory is proposed, and we use the SOINN (Self-Organizing Incremental Neural Network) method to accomplish a high accuracy abnormal detection. In the results, our method is able to shirk the video size to 10% storage size of the original video, and achieves 95% accuracy of abnormal event detection, which shows our method is useful and applicable to the surveillance industry.     

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。     

运动目标三维轨迹可视化与关联分析方法

随着治安监控系统的普及,越来越多的监控摄像头被安装在各个交通道路和公共场所中,每天都产生大量的监控视频.如今,监控视频分析工作主要是采用人工观看的方式来排查异常,以这种方式来分析视频内容耗费大量的人力和时间.目前,关于视频分析方面的研究大多是针对目标个体的异常行为检测和追踪,缺乏针对对象之间的关联关系的分析,对视频中的一些对象和场景之间的关联关系等还没有较为有效的表示和分析方法.针对这一现状,提出一种基于运动目标三维轨迹的关联视频可视分析方法来辅助人工分析视频,首先对视频资料进行预处理,获取各个目标对象的运动轨迹信息,由于二维轨迹难以处理轨迹的自相交、循环运动和停留等现象,并且没有时间信息就难以对同一空间内多个对象轨迹进行的关联性分析,于是结合时间维度对轨迹进行三维化扩展.该方法支持草图交互方式来操作,在分析过程中进行添加草图注释来辅助分析.可结合场景和对象的时空关系对轨迹进行关联性计算,得出对象及场景之间的关联模型,通过对对象在各个场景出现状况的统计,结合人工预先设定的规则,可实现对异常行为报警,辅助用户决策.     

MACHINE LEARNING TECHNIQUES FOR ACQUIRING NEW KNOWLEDGE IN IMAGE TRACKING

The purpose of this research is to apply data mining (DM) to an optimized surveillance video system with the objective of improving tracking robustness and stability. Specifically, the machine learning has been applied to blob extraction and detection, in order to decide whether a detected blob corresponds to a real target or not. Performance is assessed with an Evaluation function, which has been developed for optimizing the video surveillance system. This Evaluation function measures the quality level reached by the tracking system.     

基于手部轨迹识别的ATM智能视频监控系统

为实时阻止针对自动取款机的犯罪行为发生,设计一种基于手部轨迹识别的ATM智能视频监控系统。对于采集所得的监控区域内的视频图像,利用混合高斯背景建模方法为视频图像建立背景模型,通过背景剪除法和跟踪算法得到监控区域内的人体信息,分析进入监控区域的人体面积变化情况,由此判断是否有异常行为发生,存在异常则报警,否则采用基于颜色空间的皮肤检测算法和位置约束检测人手部分,利用隐马尔可夫模型对分段的手部运动轨迹分别进行匹配识别,进一步判断是否存在犯罪行为。实验结果表明,该方法对于犯罪行为的识别率能达到88%。     

Crowd behaviors analysis and abnormal detection based on surveillance data

Crowd analysis and abnormal trajectories detection are hot topics in computer vision and pattern recognition. As more and more video monitoring equipments are installed in public places for public security and management, researches become urgent to learn the crowd behavior patterns through the trajectories obtained by the intelligent video surveillance technology. In this paper, the FCM (Fuzzy c-means) algorithm is adopted to cluster the source points and sink points of trajectories that are deemed as critical points into several groups, and then the trajectory clusters can be acquired. The feature information statistical histogram for each trajectory cluster which contains the motion information will be built after refining them with Hausdorff distances. Eventually, the local motion coherence between test trajectories and refined trajectory clusters will be used to judge whether they are abnormal.     

Abnormal visual event detection based on multi-instance learning and autoregressive integrated moving average model in edge-based Smart City surveillance

The abnormal visual event detection is an important subject in Smart City surveillance where a lot of data can be processed locally in edge computing environment. Real-time and detection effectiveness are critical in such an edge environment. In this paper, we propose an abnormal event detection approach based on multi-instance learning and autoregressive integrated moving average model for video surveillance of crowded scenes in urban public places, focusing on real-time and detection effectiveness. We propose an unsupervised method for abnormal event detection by combining multi-instance visual feature selection and the autoregressive integrated moving average model. In the proposed method, each video clip is modeled as a visual feature bag containing several subvideo clips, each of which is regarded as an instance. The time-transform characteristics of the optical flow characteristics within each subvideo clip are considered as a visual feature instance, and time-series modeling is carried out for multiple visual feature instances related to all subvideo clips in a surveillance video clip. The abnormal events in each surveillance video clip are detected using the multi-instance fusion method. This approach is verified on publically available urban surveillance video datasets and compared with state-of-the-art alternatives. Experimental results demonstrate that the proposed method has better abnormal event detection performance for crowded scene of urban public places with an edge environment.     

Deep Event Learning boosT-up Approach: DELTA

Nowadays, the video surveillance systems may be omnipresent, but essential for supervision everywhere, e.g., ATM, airport, railway station and other crowded situations. In the multi-view video systems, various cameras are producing a huge amount of video content around the clock which makes it difficult for fast browsing, retrieval, and analysis. Accessing and managing such huge data in real time becomes a real challenging task because of inter-view dependencies, illumination changes and the bearing of many inactive frames. The work highlights an accurate and efficient technique to detect and summarize the event in multi-view surveillance videos using boosting, a machine learning algorithm, as a solution to the above issues. Interview dependencies across multiple views of the video are captured via weak learning classifiers in boosting algorithm. The light changes and still frames are tackled with moving an object in the frame by Deep learning framework. It helps to reach the correct decision for the active frame and inactive frame, without any prior information about the number of issues in a video. Target, as well as subjective ratings, clearly indicate the potency of our proposed DELTA model, where it successfully reduces the video data, while keeping the important information as events.     

Detecting rare events using Kullback–Leibler divergence: A weakly supervised approach

Video surveillance infrastructure has been widely installed in public places for security purposes. However, live video feeds are typically monitored by human staff, making the detection of important events as they occur difficult. As such, an expert system that can automatically detect events of interest in surveillance footage is highly desirable. Although a number of approaches have been proposed, they have significant limitations: supervised approaches, which can detect a specific event, ideally require a large number of samples with the event spatially and temporally localised; while unsupervised approaches, which do not require this demanding annotation, can only detect whether an event is abnormal and not specific event types. To overcome these problems, we formulate a weakly-supervised approach using Kullback–Leibler (KL) divergence to detect rare events. The proposed approach leverages the sparse nature of the target events to its advantage, and we show that this data imbalance guarantees the existence of a decision boundary to separate samples that contain the target event from those that do not. This trait, combined with the coarse annotation used by weakly supervised learning (that only indicates approximately when an event occurs), greatly reduces the annotation burden while retaining the ability to detect specific events. Furthermore, the proposed classifier requires only a decision threshold, simplifying its use compared to other weakly supervised approaches. We show that the proposed approach outperforms state-of-the-art methods on a popular real-world traffic surveillance dataset, while preserving real time performance.     

基于CNN的监控视频事件检测

复杂监控视频中事件检测是一个具有挑战性的难题, 而TRECVID-SED评测使用的数据集取自机场的实际监控视频,以高难度著称. 针对TRECVID-SED评测集, 提出了一种基于卷积神经网络(Convolutional neural network, CNN)级联网络和轨迹分析的监控视频事件检测综合方案. 在该方案中, 引入级联CNN网络在拥挤场景中准确地检测行人, 为跟踪行人奠定了基础; 采用CNN网络检测具有关键姿态的个体事件, 引入轨迹分析方法检测群体事件. 该方案在国际评测中取得了很好的评测排名: 在6个事件检测的评测中, 3个事件检测排名第一.     

Real‐time smart surveillance using motion analysis

Abstract: In the last years, smart surveillance has been one of the most active research topics in computer vision because of the wide spectrum of promising applications. Its main point is about the use of automatic video analysis technologies for surveillance purposes. In general, a processing framework for smart surveillance consists of a preliminary motion detection step in combination with high‐level reasoning that allows automatic understanding of evolutions of observed scenes. In this paper, we propose a surveillance framework based on a set of reliable visual algorithms that perform different tasks: a motion analysis approach that segments foreground regions is followed by three procedures, which perform object tracking, homographic transformations and edge matching, in order to achieve the real‐time monitoring of forbidden areas and the detection of abandoned or removed objects. Several experiments have been performed on different real image sequences acquired from a Messapic museum (indoor context) and the nearby archaeological site (outdoor context) to demonstrate the effectiveness and the flexibility of the proposed approach.     

Nonlinear control with integral sliding properties for circular aerial robot trajectory tracking: Real‐time validation

A nonlinear control algorithm for tracking dynamic trajectories using an aerial vehicle is developed in this work. The control structure is designed using a sliding mode methodology, which contains integral sliding properties. The stability analysis of the closed‐loop system is proved using the Lyapunov formalism, ensuring convergence in a desired finite time and robustness toward unknown and external perturbations from the first time instant, even for high frequency disturbances. In addition, a dynamic trajectory is constructed with the translational dynamics of an aerial robot for autonomous take‐off, surveillance missions, and landing. This trajectory respects the constraints imposed by the vehicle characteristics, allowing free initial trajectory conditions. Simulation results demonstrate the good performance of the controller in closed‐loop system when a quadrotor follows the designed trajectory. In addition, flight tests are developed to validate the trajectory and the controller behavior in real time.     

Video traffic analytics for large scale surveillance

The video traffic analysis is the most important issue for large scale surveillance. In the large scale surveillance system, huge amount of live digital video data is submitted to the storage servers through the number of externally connected scalable components. The system also contains huge amount of popular and unpopular old videos in the archived storage servers. The video data is delivered to the viewers, partly or completely on demand through a compact system. In real time, huge amount of video data is imported to the viewer’s node for various analysis purposes. The viewers use a number of interactive operations during the real time tracking suspect. The compact video on demand system is used in peer to peer mesh type hybrid architecture. The chunk of video objects move fast through the real time generated compact topological space. Video traffic analytics is required to transfer compressed multimedia data efficiently. In this work, we present a dynamically developed topological space, using mixed strategy by game approach to move the video traffic faster. The simulation results are well addressed in real life scenario.     

A real-time temporal Bayesian architecture for event surveillance and its application to patient-specific multiple disease outbreak detection

Reliable and accurate detection of disease outbreaks remains an important research topic in disease outbreak surveillance. A temporal surveillance system bases its analysis on data not only from the most recent time period, but also on data from previous time periods. A non-temporal system only looks at data from the most recent time period. There are two difficulties with a non-temporal system when it is used to monitor real data which often contain noise. First, it is prone to produce false positive signals during non-outbreak time periods. Second, during an outbreak, it tends to release false negative signals early in the outbreak, which can adversely affect the decision making process of the user of the system. We conjecture that by converting a non-temporal system to a temporal one, we may attenuate these difficulties inherent in a non-temporal system. In this paper, we propose a Bayesian network architecture for a class of temporal event surveillance models called BayesNet-T. Using this Bayesian network architecture, we can convert certain non-temporal surveillance systems to temporal ones. We apply this architecture to a previously developed non-temporal multiple-disease outbreak detection system called PC and create a temporal system called PCT. PCT takes Emergency Department (ED) patient chief complaint data as its input. The PCT system was constructed using both data (non-outbreak diseases) and expert assessments (outbreak diseases). We compare PCT to PC using a real influenza outbreak. Furthermore, we compare PCT to both PC and the classic statistical methods CUSUM and EWMA using a total of 240 influenza and Cryptosporidium disease outbreaks created by injecting stochastically simulated outbreak cases into real ED admission data. Our results indicate that PCT has a smaller mean time to detection than PC at low false alarm rates, and that PCT is more stable than PC in that once an outbreak is detected, PCT is better at maintaining the detection signal on future days.     

智能视觉监控技术研究进展

新一代智能视觉监控技术的研究是一个极具挑战性的前沿课题,它旨在赋予监控系统观察分析场景内容的能力,实现监控的自动化和智能化,因而具有巨大的应用潜力。视觉监控系统的智能化分析过程由运动目标检测、分类、跟踪和视频内容分析等几个基本环节组成,其中视频内容分析又包括异常检测、人的身份识别以及视频内容理解描述等。本文在总结以上有关关键技术研究进展的基础上,进一步提出将超分辨率复原技术引入视觉监控领域,介绍了超分辨率复原的主要算法及其在智能视觉监控中的应用。     

Hierarchical database for a multi-camera surveillance system

This paper presents a framework for event detection and video content analysis for visual surveillance applications. The system is able to coordinate the tracking of objects between multiple camera views, which may be overlapping or non-overlapping. The key novelty of our approach is that we can automatically learn a semantic scene model for a surveillance region, and have defined data models to support the storage of tracking data with different layers of abstraction into a surveillance database. The surveillance database provides a mechanism to generate video content summaries of objects detected by the system across the entire surveillance region in terms of the semantic scene model. In addition, the surveillance database supports spatio-temporal queries, which can be applied for event detection and notification applications.     

A real-time object detecting and tracking system for outdoor night surveillance

Autonomous video surveillance and monitoring has a rich history. Many deployed systems are able to reliably track human motion in indoor and controlled outdoor environments. However, object detection and tracking at night remain very important problems for visual surveillance. The objects are often distant, small and their signatures have low contrast against the background. Traditional methods based on the analysis of the difference between successive frames and a background frame will do not work. In this paper, a novel real time object detection algorithm is proposed for night-time visual surveillance. The algorithm is based on contrast analysis. In the first stage, the contrast in local change over time is used to detect potential moving objects. Then motion prediction and spatial nearest neighbor data association are used to suppress false alarms. Experiments on real scenes show that the algorithm is effective for night-time object detection and tracking.     

Illumination normalization with time-dependent intrinsic images for video surveillance

Variation in illumination conditions caused by weather, time of day, etc., makes the task difficult when building video surveillance systems of real world scenes. Especially, cast shadows produce troublesome effects, typically for object tracking from a fixed viewpoint, since it yields appearance variations of objects depending on whether they are inside or outside the shadow. In this paper, we handle such appearance variations by removing shadows in the image sequence. This can be considered as a preprocessing stage which leads to robust video surveillance. To achieve this, we propose a framework based on the idea of intrinsic images. Unlike previous methods of deriving intrinsic images, we derive time-varying reflectance images and corresponding illumination images from a sequence of images instead of assuming a single reflectance image. Using obtained illumination images, we normalize the input image sequence in terms of incident lighting distribution to eliminate shadowing effects. We also propose an illumination normalization scheme which can potentially run in real time, utilizing the illumination eigenspace, which captures the illumination variation due to weather, time of day, etc., and a shadow interpolation method based on shadow hulls. This paper describes the theory of the framework with simulation results and shows its effectiveness with object tracking results on real scene data sets.     

Activity classification and anomaly detection using m-mediods based modelling of motion patterns

Techniques for video object motion analysis, behaviour recognition and event detection are becoming increasingly important with the rapid increase in demand for and deployment of video surveillance systems. Motion trajectories provide rich spatiotemporal information about an object's activity. This paper presents a novel technique for classification of motion activity and anomaly detection using object motion trajectory. In the proposed motion learning system, trajectories are treated as time series and modelled using modified DFT-based coefficient feature space representation. A modelling technique, referred to as m-mediods, is proposed that models the class containing n members with m mediods. Once the m-mediods based model for all the classes have been learnt, the classification of new trajectories and anomaly detection can be performed by checking the closeness of said trajectory to the models of known classes. A mechanism based on agglomerative approach is proposed for anomaly detection. Four anomaly detection algorithms using m-mediods based representation of classes are proposed. These includes: (i)global merged anomaly detection (GMAD), (ii) localized merged anomaly detection (LMAD), (iii) global un-merged anomaly detection (GUAD), and (iv) localized un-merged anomaly detection (LUAD). Our proposed techniques are validated using variety of simulated and complex real life trajectory datasets.     

一种保证初始数据完整性的实时视频流分发插件

在视频监控联网系统中进行实时监控时,如果监控客户端收到的初始视频数据不完整会出现初始播放画面花屏的问题.基于开源多媒体框架GStreamer,针对H.264编码标准,设计并实现了一种实时视频流分发插件.该插件使用随机创建的请求型source衬垫用于视频数据的分发,并通过缓存当前IDR帧组的方法确保发送初始视频数据的完整性.该插件可应用于流媒体服务器当中,解决实时监控时初始画面花屏的问题.实验结果表明,应用该插件的流媒体服务器能够高效分发实时视频数据并保证初始播放画面完整,对提升实时监控效果有着明显的作用.