Abnormal crowd behavior detection using size-adapted spatio-temporal features

Abnormal crowd behavior detection is an important research issue in computer vision. However, complex real-life situations (e.g., severe occlusion, over-crowding, etc.) still challenge the effectiveness of previous algorithms. Recently, the methods based on spatio-temporal cuboid are popular in video analysis. To our knowledge, the spatio-temporal cuboid is always extracted randomly from a video sequence in the existing methods. The size of each cuboid and the total number of cuboids are determined empirically. The extracted features either contain the redundant information or lose a lot of important information which extremely affect the accuracy. In this paper, we propose an improved method. In our method, the spatio-temporal cuboid is no longer determined arbitrarily, but by the information contained in the video sequence. The spatio-temporal cuboid is extracted from video sequence with adaptive size. The total number of cuboids and the extracting positions can be determined automatically. Moreover, to compute the similarity between two spatio-temporal cuboids with different sizes, we design a novel data structure of codebook which is constructed as a set of two-level trees. The experiment results show that the detection rates of false positive and false negative are significantly reduced. Keywords: Codebook, latent dirichlet allocation (LDA), social force model, spatio-temporal cuboid.     

Abnormal crowd behavior detection using high-frequency and spatio-temporal features

Abnormal crowd behavior detection is an important research issue in computer vision. The traditional methods first extract the local spatio-temporal cuboid from video. Then the cuboid is described by optical flow or gradient features, etc. Unfortunately, because of the complex environmental conditions, such as severe occlusion, over-crowding, etc., the existing algorithms cannot be efficiently applied. In this paper, we derive the high-frequency and spatio-temporal (HFST) features to detect the abnormal crowd behaviors in videos. They are obtained by applying the wavelet transform to the plane in the cuboid which is parallel to the time direction. The high-frequency information characterize the dynamic properties of the cuboid. The HFST features are applied to the both global and local abnormal crowd behavior detection. For the global abnormal crowd behavior detection, Latent Dirichlet allocation is used to model the normal scenes. For the local abnormal crowd behavior detection, Multiple Hidden Markov Models, with an competitive mechanism, is employed to model the normal scenes. The comprehensive experiment results show that the speed of detection has been greatly improved using our approach. Moreover, a good accuracy has been achieved considering the false positive and false negative detection rates.     

基于时空大数据的异常车辆检测

针对传统人工检查黑车的方式不但耗时耗力而且效率低下的问题,提出一种新的自动检测黑车的方法。在Hadoop平台上,对物联网技术采集的全疆车辆加气数据进行分析;抽取车辆加气的时间特征和空间特征;利用随机森林算法研究车辆与驾驶员、加气站间的关系,从而发现具有异常加气模式的黑车车辆。在大规模真实数据集上的实验表明:提出的方法在黑车发现问题上有较高的准确率,可以用于帮助有关部门提高黑车检测的效率。     

无偏KL散度算法对时空异常区间检测的优化研究

通过对多变量时空时间序列中异常的度量,可以从大量时空事件数据中检测出异常的数据部分。与孤立异常数据点检测采用的技术不同,提出了无偏KL散度算法(UKLD)。首先定义了时空时间序列中的异常区间,嵌入时间延迟后用高斯分布来估计检测区间和剩余区间的分布并通过累计和来加快高斯分布的参数估计过程,最后使用无偏KL散度计算区间之间的差异水平,将这种差异水平作为检测区间的异常得分从而得到时空异常区间。仿真分析结果表明,对比HOT SAX算法和RKDE算法,UKLD算法在精度方面更优,能更好地实现时空数据中的异常区间检测。     

Bio-inspired feature enhancement network for edge detection

As the basis of mid-level and high-level vision tasks, edge detection has great significance in the field of computer vision. Edge detection methods based on deep learning usually adopt the structure of the encoding-decoding network, among which the deep convolutional neural network is generally adopted in the encoding network, and the decoding network is designed by researchers. In the design of the encoding-decoding network, researchers pay more attention to the design of the decoding network and ignore the influence of the encoding network, which makes the existing edge detection methods have the problems of weak feature extraction ability and insufficient edge information extraction. To improve the existing methods, this work combines the information transmission mechanism of the retina/lateral geniculate nucleus with an edge detection network based on convolutional neural network and proposes a bionic feature enhancement network. It consists of a pre-enhanced network, an encoding network, and a decoding network. By simulating the information transfer mechanism of the retina/lateral geniculate nucleus, we designed the pre-enhanced network to enhance the ability of the encoding network to extract details and local features. Based on the hierarchical structure of the visual pathway and the integrated feature function of the inferior temporal (IT) cortex, we designed a novel feature fusion network as a decoding network. In a feature fusion network, a down-sampling enhancement module is introduced to boost the feature integration ability of the decoding network. Experimental results demonstrate that we achieve state-of-the-art performance on several available datasets.

     

Inference of network anomaly propagation using spatio-temporal correlation

Many solutions have been proposed for network alarm correlation. However, they mainly have focused on alarm reduction and on root cause analysis. This paper presents an automated alarm correlation system composed of three layers, which obtains raw alarms and presents to network administrator a wide view of the scenario affected by the volume anomaly. In the preprocessing layer, it is performed the alarm compression using their spatial and temporal attributes, which are reduced into a unique alarm named Device Level Alarm (DLA). The correlation layer aims to infer the anomaly propagation path and its origin and destination using DLAs and network topology information. The presentation layer provides the visualization of the path and network elements affected by the anomaly propagation. Moreover, it is presented the Anomaly Propagation View (APV), a graphic tool developed to provide a wide visualization of the network status. In order to evaluate the effectiveness of the proposed solution, it was used real traffic data from State University of Londrina.     

Human body pose detection using Bayesian spatio-temporal templates

We present a template-based approach to detecting human silhouettes in a specific walking pose. Our templates consist of short sequences of 2D silhouettes obtained from motion capture data. This lets us incorporate motion information into them and helps distinguish actual people who move in a predictable way from static objects whose outlines roughly resemble those of humans. Moreover, during the training phase we use statistical learning techniques to estimate and store the relevance of the different silhouette parts to the recognition task. At run-time, we use it to convert Chamfer distance to meaningful probability estimates. The templates can handle six different camera views, excluding the frontal and back view, as well as different scales. We demonstrate the effectiveness of our technique using both indoor and outdoor sequences of people walking in front of cluttered backgrounds and acquired with a moving camera, which makes techniques such as background subtraction impractical.     

Object motion detection using information theoretic spatio-temporal saliency

This paper proposes to employ the visual saliency for moving object detection via direct analysis from videos. Object saliency is represented by an information saliency map (ISM), which is calculated from spatio-temporal volumes. Both spatial and temporal saliencies are calculated and a dynamic fusion method developed for combination. We use dimensionality reduction and kernel density estimation to develop an efficient information theoretic based procedure for constructing the ISM. The ISM is then used for detecting foreground objects. Three publicly available visual surveillance databases, namely CAVIAR, PETS and OTCBVS-BENCH are selected for evaluation. Experimental results show that the proposed method is robust for both fast and slow moving object detection under illumination changes. The average detection rates are 95.42% and 95.81% while the false detection rates are 2.06% and 2.40% in CAVIAR (INRIA entrance hall and shopping center) dataset and OTCBVS-BENCH database, respectively. The average processing speed is 6.6 fps with frame resolution 320×240 in a typical Pentium IV computer.     

一种细化边缘的轻量级边缘检测神经网络

针对目前边缘检测模型复杂度高、参数量大、识别边缘精度及效率不高的问题,提出了一种轻量级边缘检测神经网络。首先,通过MP模块将原图像分成同等维度的小块并进行位置编码,以增强边缘像素点之间的联系;再经过MGC中的多组卷积操作提取图像不同区域特征进行融合,减少冗余信息;最后通过多次的反卷积上采样调整输出图片的分辨率尺寸,输出预测边缘图。最终的网络只有125 KB的参数,在BIPED和MDBD数据集上进行实验,验证模型检测边缘的综合能力。相比于当前最先进的轻量级边缘检测方法LDC,在BIPED数据集上的测试结果表明,指标ODS（optimal dataset scale）仅低了0.9%,模型参数量则降低了81.5%,FPS提高了89.0%。在保持细粒度识别边缘的同时,可以满足实际任务中的需求。     

Color edge detection using the minimal spanning tree

In this study, the edge detection task in vector-valued images is examined as a clustering problem. Using samples within a data window, the minimal spanning tree (MST) provides the ordering of multivariate observations and facilitates the identification of similar classes. The edge detector parameters like edge strength, type and orientation are subsequently determined from the clustered data. Experiments and comparisons are performed, revealing the enhanced performance of the proposed approach.     

Abnormal behavior detection using dominant sets

Smart surveillance systems are increasingly being used to detect potentially dangerous situations. To do so, the common and easier way is to model normal human behaviors and consider as abnormal any new strange behavior in the scene. In this article, Dominant Sets is adapted to model most frequent behaviors and to detect any unknown event to trigger an alarm. It is proved that after an unsupervised training, Dominant Sets can robustly detect abnormal behaviors. The method is tested in several different cases and compared to other usual clusterization methods such as KNN, mixture of Gaussians or Fuzzy \(K\) -Means to confirm its robustness and performance. The overall performance of abnormal behavior detection based on Dominant Sets is better, being the error ratio at least \(1.5\) points lower than the others.     

Abnormal driving behavior detection based on kernelization-sparse representation in video surveillance

The detection of abnormal driving behaviors based on video surveillance systems is an important part of Intelligent Transportation System (ITS), which can help reduce disturbances on traffic flow and improve traffic safety. First, the study proposes a novel nonlinear sparse reconstruction method for abnormal driving behavior detection in video surveillance. A hybrid kernel function formed by convexly combining a local kernel of radial basis function (RBF) and a global kernel of homogeneous polynomial is been applied in sparse reconstruction method. Then, a novel Hybrid Kernel Orthogonal Matching Pursuit (HKOMP) algorithm is designed to solve the proposed sparse reconstruction model. Finally, the performance of the abnormal detection method is tested on two datasets i.e. stop sign dataset and car parking dataset. In addition, comparative experiments with five classical methods are carried out. The experimental results indicate that the proposed method outperforms other five comparison methods in terms of accuracy.

     

Occlusion event detection using geometric features in spatio-temporal volumes

In video sequences, edges in 2D images (frames) produce 3D surface in the spatio-temporal volume. In this paper, we propose to consider temporal collisions between edges, and thus objects, as 3D ridges in the spatio-temporal volume. Edge collisions (i.e. ridge points) can be located using the maximum principal curvature and the principal curvature direction. Using the detected ridges, we then propose a technique to identify overlapping objects events in an image sequence, by neither computing depth nor optical flow. We present successful experiments on real image sequences.     

Scalable architecture for an automated surveillance system using edge computing

Over the past few years, the Internet of Things has gone from theoretical concept to our everyday living experience. The explosive growth of sensor streams also leads to a new paradigm of edge computing. In the surveillance system, edge-based automation is crucial to get fast response for fast data analytics among connected devices. In this paper, we propose an automated surveillance system to improve robustness and intelligence. Our scalable architecture is an alternative way of reducing the server resource and wireless network limitation.     

BP神经网络在图像边缘检测中的应用

为实现数字图像边缘的有效检测与提取,借助BP神经网络,采用了改进的最速梯度下降法,通过对动量项的合理选择,有效地实现算法的快速收敛。为提高算法的执行效率,采用直接编程和对图像采用分块的思想,并给出了算法实现的方法和步骤。用Matlab软件对灰度图像进行了仿真,并将仿真结果和传统的方法进行了比较,结果表明,所设计的网络边缘检测优于传统方法,并具有较好的泛化能力。     

基于深度学习的海岸线边缘检测网络模型

海岸线的动态监测对海岸带的规划管理具有非常重要的意义。由于海陆环境错综复杂,遥感影像中海陆边界光谱特征不明显,导致提取的海岸线定位不准确。提出一种融合语义分割网络和边缘检测网络的深度卷积神经网络模型(EWNet)。该模型包含2个分支流：语义分割流负责提取分层语义信息并用来指导边缘检测流获取岸线语义信息;边缘检测流通过语义分割流完善边缘语义信息。在“高分一号”遥感图像上的实验结果表明,与几种最新网络模型相比,EWNet获得了更精确的海岸线边界提取结果。     

Optimal ramp edge detection using expansion matching

In practical images, ideal step edges are actually transformed into ramp edges, due to the general low pass filtering nature of imaging systems. This paper discusses the application of the expansion matching (EXM) method for optimal ramp edge detection. EXM optimizes a novel matching criterion called discriminative signal-to-noise ratio (DSNR) and has been shown to robustly recognize templates under conditions of noise, severe occlusion, and superposition. We show that our ramp edge detector performs better than the ramp detector obtained from Canny's criteria in terms of DSNR and is relatively easier to derive for various noise levels and slopes     

A new gravitational image edge detection method using edge explorer agents

In this paper, a new algorithm for image edge detection based on the theory of universal gravity is proposed. The problem is represented by a discrete space in which each image pixel is considered as a celestial body and its mass is considered to be corresponding to the pixel’s grayscale intensity. To find the edgy pixels a number of moving agents are randomly generated and initialized through the image space. Artificial agents move through the space via the forces of celestial bodies that are located in their neighborhood and in this way they can find the promising edge pixels. A large number of experiments are employed to determine suitable algorithm parameters and confirm the legitimacy of the proposed algorithm. Also, the results are compared with conventional and soft computing based methods like Sobel, Canny and ant-based edge detector. As compared to other standard techniques, our algorithm provides more accurate results over 11 test images via Baddeley’s error metric. The visual and quantitative comparisons reveal the effectiveness and robustness of the proposed algorithm.     

Vehicle re-identification using multi-task deep learning network and spatio-temporal model

Multimedia Tools and Applications - Vehicle re-identification (re-ID) plays an important role in the automatic analysis of the increasing urban surveillance videos and has become a hot topic in...     

Image filtering, edge detection, and edge tracing using fuzzyreasoning

We characterize the problem of detecting edges in images as a fuzzy reasoning problem. The edge detection problem is divided into three stages: filtering, detection, and tracing. Images are filtered by applying fuzzy reasoning based on local pixel characteristics to control the degree of Gaussian smoothing. Filtered images are then subjected to a simple edge detection algorithm which evaluates the edge fuzzy membership value for each pixel, based on local image characteristics. Finally, pixels having high edge membership are traced and assembled into structures, again using fuzzy reasoning to guide the tracing process. The filtering, detection, and tracing algorithms are tested on several test images. Comparison is made with a standard edge detection technique