 |
Video Surveillance |
Video surveillance is a fast growing area of public security. With it has come policy issues related to privacy. Technical issues and opportunities have also arisen, including the potential to use advanced methods to provide positive identification, abnormal behaviors in crowds, intruder detection, and information fusion with other data. The research presented here came from multiple conferences and publications and was offered in 2014.
Xiaochun Cao; Na Liu; Ling Du; Chao Li, "Preserving Privacy For Video Surveillance Via Visual Cryptography," Signal and Information Processing (ChinaSIP), 2014 IEEE China Summit & International Conference on, pp.607,610, 9-13 July 2014. doi: 10.1109/ChinaSIP.2014.6889315 The video surveillance widely installed in public areas poses a significant threat to the privacy. This paper proposes a new privacy preserving method via the Generalized Random-Grid based Visual Cryptography Scheme (GRG-based VCS). We first separate the foreground from the background for each video frame. These foreground pixels contain the most important information that needs to be protected. Every foreground area is encrypted into two shares based on GRG-based VCS. One share is taken as the foreground, and the other one is embedded into another frame with random selection. The content of foreground can only be recovered when these two shares are got together. The performance evaluation on several surveillance scenarios demonstrates that our proposed method can effectively protect sensitive privacy information in surveillance videos.
Keywords: cryptography; data protection; video surveillance; GRG-based VCS; foreground pixels; generalized random-grid based visual cryptography scheme; performance evaluation; random selection; sensitive privacy information preservation method; video frame; video surveillance; Cameras; Cryptography; PSNR; Privacy; Video surveillance; Visualization; Random-Grid; Video surveillance; privacy protection; visual cryptography (ID#: 15-3584)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6889315&isnumber=6889177
Yoohwan Kim; Juyeon Jo; Shrestha, S., "A Server-Based Real-Time Privacy Protection Scheme Against Video Surveillance By Unmanned Aerial Systems," Unmanned Aircraft Systems (ICUAS), 2014 International Conference on, pp.684,691, 27-30 May 2014. doi: 10.1109/ICUAS.2014.6842313 Abstract: Unmanned Aerial Systems (UAS) have raised a great concern on privacy recently. A practical method to protect privacy is needed for adopting UAS in civilian airspace. This paper examines the privacy policies, filtering strategies, existing techniques, then proposes a novel method based on the encrypted video stream and the cloud-based privacy servers. In this scheme, all video surveillance images are initially encrypted, then delivered to a privacy server. The privacy server decrypts the video using the shared key with the camera, and filters the image according to the privacy policy specified for the surveyed region. The sanitized video is delivered to the surveillance operator or anyone on the Internet who is authorized. In a larger system composed of multiple cameras and multiple privacy servers, the keys can be distributed using Kerberos protocol. With this method the privacy policy can be changed on demand in real-time and there is no need for a costly on-board processing unit. By utilizing the cloud-based servers, advanced image processing algorithms and new filtering algorithms can be applied immediately without upgrading the camera software. This method is cost-efficient and promotes video sharing among multiple subscribers, thus it can spur wide adoption.
Keywords: Internet; data privacy; video coding; video surveillance; Internet; Kerberos protocol; UAS; camera software; civilian airspace; cloud-based privacy servers; cloud-based servers; encrypted video stream; filtering algorithms; filtering strategies; image processing algorithms; multiple privacy servers; on-board processing unit; privacy policy; sanitized video; server-based real-time privacy protection scheme; surveillance operator; unmanned aerial systems; video sharing; video surveillance images; Cameras; Cryptography; Filtering; Privacy; Servers; Streaming media; Surveillance; Key Distribution; Privacy; Unmanned Aerial Systems; Video Surveillance (ID#: 15-3585)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6842313&isnumber=6842225
Hassan, M.M.; Hossain, M.A.; Al-Qurishi, M., "Cloud-based Mobile IPTV Terminal for Video Surveillance," Advanced Communication Technology (ICACT), 2014 16th International Conference on, pp.876, 880, 16-19 Feb. 2014. doi: 10.1109/ICACT.2014.6779086 Surveillance video streams monitoring is an important task that the surveillance operators usually carry out. The distribution of video surveillance facilities over multiple premises and the mobility of surveillance users requires that they are able to view surveillance video seamlessly from their mobile devices. In order to satisfy this requirement, we propose a cloud-based IPTV (Internet Protocol Television) solution that leverages the power of cloud infrastructure and the benefits of IPTV technology to seamlessly deliver surveillance video content on different client devices anytime and anywhere. The proposed mechanism also supports user-controlled frame rate adjustment of video streams and sharing of these streams with other users. In this paper, we describe the overall approach of this idea, address and identify key technical challenges for its practical implementation. In addition, initial experimental results were presented to justify the viability of the proposed cloud-based IPTV surveillance framework over the traditional IPTV surveillance approach.
Keywords: IPTV; cloud computing; mobile television; video surveillance Internet protocol television ;cloud-based mobile IPTV terminal; mobile devices; surveillance operators; surveillance video streams monitoring; video surveillance facilities distribution; Cameras ;IPTV; Mobile communication; Servers; Streaming media; Video surveillance; IPTV; Video surveillance; cloud computing; mobile terminal (ID#: 15-3586)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6779086&isnumber=6778899
Gorur, P.; Amrutur, B., "Skip Decision and Reference Frame Selection for Low-Complexity H.264/AVC Surveillance Video Coding," Circuits and Systems for Video Technology, IEEE Transactions on, vol.24, no.7, pp.1156,1169, July 2014. doi: 10.1109/TCSVT.2014.2319611 H.264/advanced video coding surveillance video encoders use the Skip mode specified by the standard to reduce bandwidth. They also use multiple frames as reference for motion-compensated prediction. In this paper, we propose two techniques to reduce the bandwidth and computational cost of static camera surveillance video encoders without affecting detection and recognition performance. A spatial sampler is proposed to sample pixels that are segmented using a Gaussian mixture model. Modified weight updates are derived for the parameters of the mixture model to reduce floating point computations. A storage pattern of the parameters in memory is also modified to improve cache performance. Skip selection is performed using the segmentation results of the sampled pixels. The second contribution is a low computational cost algorithm to choose the reference frames. The proposed reference frame selection algorithm reduces the cost of coding uncovered background regions. We also study the number of reference frames required to achieve good coding efficiency. Distortion over foreground pixels is measured to quantify the performance of the proposed techniques. Experimental results show bit rate savings of up to 94.5% over methods proposed in literature on video surveillance data sets. The proposed techniques also provide up to 74.5% reduction in compression complexity without increasing the distortion over the foreground regions in the video sequence.
Keywords: Gaussian processes; cameras; data compression; distortion; motion compensation; video codecs; video coding; video surveillance; Gaussian mixture model;H.264/advanced video coding surveillance video encoders; bit rate savings; coding uncovered background regions; compression complexity; detection performance; distortion; floating point computations; foreground pixels; low-complexity H.264/AVC surveillance video coding; mixture model; motion-compensated prediction; multiple frames; recognition performance; reference frame selection; reference frame selection algorithm; skip decision; static camera surveillance video encoders; video sequence; video surveillance data sets; Cameras; Encoding; Motion detection; Motion segmentation; Streaming media; Surveillance; Video coding; Cache optimization; H.264/advanced video coding (AVC); motion detection; reference frame selection; skip decision; video surveillance (ID#: 15-3587)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6805578&isnumber=6846390
Xianguo Zhang; Tiejun Huang; Yonghong Tian; Wen Gao, "Background-Modeling-Based Adaptive Prediction for Surveillance Video Coding," Image Processing, IEEE Transactions on, vol.23, no.2, pp.769,784, Feb. 2014. doi: 10.1109/TIP.2013.2294549 The exponential growth of surveillance videos presents an unprecedented challenge for high-efficiency surveillance video coding technology. Compared with the existing coding standards that were basically developed for generic videos, surveillance video coding should be designed to make the best use of the special characteristics of surveillance videos (e.g., relative static background). To do so, this paper first conducts two analyses on how to improve the background and foreground prediction efficiencies in surveillance video coding. Following the analysis results, we propose a background-modeling-based adaptive prediction (BMAP) method. In this method, all blocks to be encoded are firstly classified into three categories. Then, according to the category of each block, two novel inter predictions are selectively utilized, namely, the background reference prediction (BRP) that uses the background modeled from the original input frames as the long-term reference and the background difference prediction (BDP) that predicts the current data in the background difference domain. For background blocks, the BRP can effectively improve the prediction efficiency using the higher quality background as the reference; whereas for foreground-background-hybrid blocks, the BDP can provide a better reference after subtracting its background pixels. Experimental results show that the BMAP can achieve at least twice the compression ratio on surveillance videos as AVC (MPEG-4 Advanced Video Coding) high profile, yet with a slightly additional encoding complexity. Moreover, for the foreground coding performance, which is crucial to the subjective quality of moving objects in surveillance videos, BMAP also obtains remarkable gains over several state-of-the-art methods.
Keywords: data compression; video coding; video surveillance; AVC; BDP; BMAP method; BRP; MPEG-4 advanced video coding; background difference prediction; background pixels; background prediction efficiency; background reference prediction; background-modeling-based adaptive prediction method; encoding complexity; exponential growth; foreground coding performance; foreground prediction efficiency; foreground-background-hybrid blocks; high-efficiency surveillance video coding technology; surveillance video compression ratio; Complexity theory; Decoding; Encoding; Image coding; Object oriented modeling; Surveillance; video coding; Surveillance video; background difference; background modeling; background reference; block classification (ID#: 15-3588)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6680670&isnumber=6685907
Chun-Rong Huang; Chung, P.-C.J.; Di-Kai Yang; Hsing-Cheng Chen; Guan-Jie Huang, "Maximum a Posteriori Probability Estimation for Online Surveillance Video Synopsis," Circuits and Systems for Video Technology, IEEE Transactions on, vol.24, no.8, pp.1417,1429, Aug. 2014. doi: 10.1109/TCSVT.2014.2308603 To reduce human efforts in browsing long surveillance videos, synopsis videos are proposed. Traditional synopsis video generation applying optimization on video tubes is very time consuming and infeasible for real-time online generation. This dilemma significantly reduces the feasibility of synopsis video generation in practical situations. To solve this problem, the synopsis video generation problem is formulated as a maximum a posteriori probability (MAP) estimation problem in this paper, where the positions and appearing frames of video objects are chronologically rearranged in real time without the need to know their complete trajectories. Moreover, a synopsis table is employed with MAP estimation to decide the temporal locations of the incoming foreground objects in the synopsis video without needing an optimization procedure. As a result, the computational complexity of the proposed video synopsis generation method can be significantly reduced. Furthermore, as it does not require prescreening the entire video, this approach can be applied on online streaming videos.
Keywords: maximum likelihood estimation; video signal processing; video streaming; video surveillance; MAP estimation problem; computational complexity reduction; human effort reduction; long surveillance video browsing; maximum-a-posteriori probability estimation problem; online streaming videos; online surveillance video synopsis; synopsis table; synopsis video generation problem; video summarization; video tubes; Estimation; Indexes; Optimization; Predictive models; Real-time systems; Streaming media; Surveillance; Maximum a posteriori (MAP) estimation; video summarization; video surveillance; video synopsis (ID#: 15-3589)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6748870&isnumber=6869080
Hong Jiang; Songqing Zhao; Zuowei Shen; Wei Deng; Wilford, P.A.; Haimi-Cohen, R., "Surveillance Video Analysis Using Compressive Sensing With Low Latency," Bell Labs Technical Journal , vol.18, no.4, pp. 63, 74, March 2014. doi: 10.1002/bltj.21646 We propose a method for analysis of surveillance video by using low rank and sparse decomposition (LRSD) with low latency combined with compressive sensing to segment the background and extract moving objects in a surveillance video. Video is acquired by compressive measurements, and the measurements are used to analyze the video by a low rank and sparse decomposition of a matrix. The low rank component represents the background, and the sparse component, which is obtained in a tight wavelet frame domain, is used to identify moving objects in the surveillance video. An important feature of the proposed low latency method is that the decomposition can be performed with a small number of video frames, which reduces latency in the reconstruction and makes it possible for real time processing of surveillance video. The low latency method is both justified theoretically and validated experimentally.
Keywords: compressed sensing; image motion analysis; image segmentation; video surveillance; wavelet transforms; LRSD; background segmentation; compressive sensing; low latency method; low rank and sparse decomposition; surveillance video analysis; video frames; wavelet frame domain; Matrix decompoistion; Object recognition; Sparse decomposition; Sparse matrices; Streaming media; Surveillance; Video communication; Wavelet domain (ID#: 15-3590)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6770348&isnumber=6770341
Rasheed, N.; Khan, S.A.; Khalid, A., "Tracking and Abnormal Behavior Detection in Video Surveillance Using Optical Flow and Neural Networks," Advanced Information Networking and Applications Workshops (WAINA), 2014 28th International Conference on, pp.61,66, 13-16 May 2014. doi: 10.1109/WAINA.2014.18 An abnormal behavior detection algorithm for surveillance is required to correctly identify the targets as being in a normal or chaotic movement. A model is developed here for this purpose. The uniqueness of this algorithm is the use of foreground detection with Gaussian mixture (FGMM) model before passing the video frames to optical flow model using Lucas-Kanade approach. Information of horizontal and vertical displacements and directions associated with each pixel for object of interest is extracted. These features are then fed to feed forward neural network for classification and simulation. The study is being conducted on the real time videos and some synthesized videos. Accuracy of method has been calculated by using the performance parameters for Neural Networks. In comparison of plain optical flow with this model, improved results have been obtained without noise. Classes are correctly identified with an overall performance equal to 3.4e-02 with & error percentage of 2.5.
Keywords: Gaussian processes; feature selection; feedforward neural nets; image sequences; mixture models; object detection; video surveillance; FGMM model; Lucas-Kanade approach; abnormal behavior detection; chaotic movement; feed forward neural network; foreground detection with Gaussian mixture model; neural networks; normal movement; optical flow; real time videos; synthesized videos; targets identification; video frames; video surveillance; Adaptive optics; Computer vision; Image motion analysis; Neural networks; Optical computing; Optical imaging; Streaming media; Foreground Detection; Gaussian Mixture Models; Neural Network; Optical Flow; Video Surveillance (ID#: 15-3591)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6844614&isnumber=6844560
Hammoud, R.I.; Sahin, C.S.; Blasch, E.P.; Rhodes, B.J., "Multi-source Multi-modal Activity Recognition in Aerial Video Surveillance," Computer Vision and Pattern Recognition Workshops (CVPRW), 2014 IEEE Conference on, pp.237,244, 23-28 June 2014. doi: 10.1109/CVPRW.2014.44 Recognizing activities in wide aerial/overhead imagery remains a challenging problem due in part to low-resolution video and cluttered scenes with a large number of moving objects. In the context of this research, we deal with two un-synchronized data sources collected in real-world operating scenarios: full-motion videos (FMV) and analyst call-outs (ACO) in the form of chat messages (voice-to-text) made by a human watching the streamed FMV from an aerial platform. We present a multi-source multi-modal activity/event recognition system for surveillance applications, consisting of: (1) detecting and tracking multiple dynamic targets from a moving platform, (2) representing FMV target tracks and chat messages as graphs of attributes, (3) associating FMV tracks and chat messages using a probabilistic graph-based matching approach, and (4) detecting spatial-temporal activity boundaries. We also present an activity pattern learning framework which uses the multi-source associated data as training to index a large archive of FMV videos. Finally, we describe a multi-intelligence user interface for querying an index of activities of interest (AOIs) by movement type and geo-location, and for playing-back a summary of associated text (ACO) and activity video segments of targets-of-interest (TOIs) (in both pixel and geo-coordinates). Such tools help the end-user to quickly search, browse, and prepare mission reports from multi-source data.
Keywords: image matching; image motion analysis; image representation; indexing; learning (artificial intelligence);object detection; query processing; target tracking; user interfaces; video streaming; video surveillance; ACO; FMV streaming; FMV target track representation; FMV videos; activities of interest; activity pattern learning framework; activity video segments; aerial imagery; aerial video surveillance; analyst call-outs; associated text; full-motion video; geolocation; index query; multi-intelligence user interface; multiple dynamic target detection; multiple dynamic target tracking; multisource associated data; multisource multimodal activity recognition; multisource multimodal event recognition; overhead imagery; probabilistic graph-based matching approach; spatial-temporal activity boundary detection; targets-of-interest; unsynchronized data sources; voice-to-text chat messages; Pattern recognition; Radar tracking; Semantics; Streaming media; Target tracking; Vehicles; FMV exploitation; MINER; activity recognition; chat and video fusion; event recognition; fusion; graph matching; graph representation; surveillance (ID#: 15-3592)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6909989&isnumber=6909944
Woon Cho; Abidi, M.A.; Kyungwon Jeong; Nahyun Kim; Seungwon Lee; Joonki Paik; Gwang-Gook Lee, "Object Retrieval Using Scene Normalized Human Model For Video Surveillance System," Consumer Electronics (ISCE 2014), The 18th IEEE International Symposium on, pp.1,2, 22-25 June 2014 doi: 10.1109/ISCE.2014.6884439 This paper presents a human model-based feature extraction method for a video surveillance retrieval system. The proposed method extracts, from a normalized scene, object features such as height, speed, and representative color using a simple human model based on multiple-ellipse. Experimental results show that the proposed system can effectively track moving routes of people such as a missing child, an absconder, and a suspect after events.
Keywords: feature extraction; image retrieval; object tracking; feature extraction; multiple ellipse human model; object retrieval; scene normalized human model; video surveillance retrieval system; Cameras; Databases; Feature extraction ;Image color analysis; Shape; Video surveillance; human model; retrieval system; scene calibration; surveillance system (ID#: 15-3593)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6884439&isnumber=6884278
Ma Juan; Hu Rongchun; Li Jian, "A Fast Human Detection Algorithm Of Video Surveillance In Emergencies," Control and Decision Conference (2014 CCDC), The 26th Chinese, vol., no., pp.1500,1504, May 31 2014-June 2 2014. doi: 10.1109/CCDC.2014.6852404 This paper propose a fast human detection algorithm of video surveillance in emergencies. Firstly through the background subtraction based on the single Guassian model and frame subtraction, we get the target mask which is optimized by Gaussian filter and dilation. Then the interest points of head is obtained from figures with target mask and edge detection. Finally according to detecting these points we can track the head and count the number of people with the frequency of moving target at the same place. Simulation results show that the algorithm can detect the moving object quickly and accurately.
Keywords: Gaussian processes; edge detection; object detection; video surveillance; Gaussian filter; background subtraction; dilation; edge detection; emergencies; frame subtraction; human detection algorithm; moving target; single Guassian model; target mask; video surveillance; Conferences; Detection algorithms; Educational institutions; Electronic mail; Estimation; IEEE Computer Society; Image edge detection; background subtraction; edge tracking of head; frame subtraction; target mask (ID#: 15-3594)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6852404&isnumber=6852105
Harish, Palagati; Subhashini, R.; Priya, K., "Intruder Detection By Extracting Semantic Content From Surveillance Videos," Green Computing Communication and Electrical Engineering (ICGCCEE), 2014 International Conference on, pp.1,5, 6-8 March 2014. doi: 10.1109/ICGCCEE.2014.6922469 Many surveillance cameras are using everywhere, the videos or images captured by these cameras are still dumped but they are not processed. Many methods are proposed for tracking and detecting the objects in the videos but we need the meaningful content called semantic content from these videos. Detecting Human activity recognition is quite complex. The proposed method called Semantic Content Extraction (SCE) from videos is used to identify the objects and the events present in the video. This model provides useful methodology for intruder detecting systems which provides the behavior and the activities performed by the intruder. Construction of ontology enhances the spatial and temporal relations between the objects or features extracted. Thus proposed system provides a best way for detecting the intruders, thieves and malpractices happening around us.
Keywords: Cameras; Feature extraction; Ontologies; Semantics; Video surveillance; Videos; Human activity recognition; Ontology; Semantic content; Spatial and Temporal Relations (ID#: 15-3595)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6922469&isnumber=6920919
Wang, S.; Orwell, J.; Hunter, G., "Evaluation of Bayesian and Dempster-Shafer Approaches to Fusion of Video Surveillance Information," Information Fusion (FUSION), 2014 17th International Conference on, pp. 1, 7, 7-10 July 2014. (no doi provided) This paper presents the application of fusion methods to a visual surveillance scenario. The range of relevant features for re-identifying vehicles is discussed, along with the methods for fusing probabilistic estimates derived from these estimates. In particular, two statistical parametric fusion methods are considered: Bayesian Networks and the Dempster Shafer approach. The main contribution of this paper is the development of a metric to allow direct comparison of the benefits of the two methods. This is achieved by generalising the Kelly betting strategy to accommodate a variable total stake for each sample, subject to a fixed expected (mean) stake. This metric provides a method to quantify the extra information provided by the Dempster-Shafer method, in comparison to a Bayesian Fusion approach.
Keywords: Accuracy; Bayes methods; Color; Mathematical model; Shape; Uncertainty; Vehicles; Bayesian; Dempster-Shafer; evaluation; fusion ;vehicle (ID#: 15-3596)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6916172&isnumber=6915967
Yueguo Zhang; Lili Dong; Shenghong Li; Jianhua Li, "Abnormal Crowd Behavior Detection Using Interest Points," Broadband Multimedia Systems and Broadcasting (BMSB), 2014 IEEE International Symposium on, pp.1,4, 25-27 June 2014. doi: 10.1109/BMSB.2014.6873527 Abnormal crowd behavior detection is an important research issue in video processing and computer vision. In this paper we introduce a novel method to detect abnormal crowd behaviors in video surveillance based on interest points. A complex network-based algorithm is used to detect interest points and extract the global texture features in scenarios. The performance of the proposed method is evaluated on publicly available datasets. We present a detailed analysis of the characteristics of the crowd behavior in different density crowd scenes. The analysis of crowd behavior features and simulation results are also demonstrated to illustrate the effectiveness of our proposed method.
Keywords: behavioural sciences computing; complex networks; computer vision; feature extraction ;image texture; object detection; video signal processing; video surveillance; abnormal crowd behavior detection; complex network-based algorithm; computer vision; crowd behavior feature analysis; global texture feature extraction; interest point detection; video processing; video surveillance; Broadband communication; Broadcasting; Complex networks; Computer vision; Feature extraction; Multimedia systems; Video surveillance; Crowd Behavior; Video Surveillance; Video processing (ID#: 15-3597)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6873527&isnumber=6873457
Lu Wang; Yung, N.H.C.; Lisheng Xu, "Multiple-Human Tracking by Iterative Data Association and Detection Update," Intelligent Transportation Systems, IEEE Transactions on, vol. 15, no. 5, pp.1886,1899, Oct. 2014. doi: 10.1109/TITS.2014.2303196 Multiple-object tracking is an important task in automated video surveillance. In this paper, we present a multiple-human-tracking approach that takes the single-frame human detection results as input and associates them to form trajectories while improving the original detection results by making use of reliable temporal information in a closed-loop manner. It works by first forming tracklets, from which reliable temporal information is extracted, and then refining the detection responses inside the tracklets, which also improves the accuracy of tracklets' quantities. After this, local conservative tracklet association is performed and reliable temporal information is propagated across tracklets so that more detection responses can be refined. The global tracklet association is done last to resolve association ambiguities. Experimental results show that the proposed approach improves both the association and detection results. Comparison with several state-of-the-art approaches demonstrates the effectiveness of the proposed approach.
Keywords: feature extraction; intelligent transportation systems; iterative methods ;object tracking; sensor fusion; video surveillance; automated video surveillance; detection responses; human detection results; intelligent transportation systems; iterative data association; multiple-human tracking; temporal information extraction ;tracklet association; Accuracy; Computational modeling; Data mining; Reliability; Solid modeling; Tracking; Trajectory; Data association; detection update; multiple-human tracking; video surveillance (ID#: 15-3598)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6750747&isnumber=6910343
Shuai Yi; Xiaogang Wang, "Profiling Stationary Crowd Groups," Multimedia and Expo (ICME), 2014 IEEE International Conference on, pp. 1, 6, 14-18 July 2014. doi: 10.1109/ICME.2014.6890138 Detecting stationary crowd groups and analyzing their behaviors have important applications in crowd video surveillance, but have rarely been studied. The contributions of this paper are in two aspects. First, a stationary crowd detection algorithm is proposed to estimate the stationary time of foreground pixels. It employs spatial-temporal filtering and motion filtering in order to be robust to noise caused by occlusions and crowd clutters. Second, in order to characterize the emergence and dispersal processes of stationary crowds and their behaviors during the stationary periods, three attributes are proposed for quantitative analysis. These attributes are recognized with a set of proposed crowd descriptors which extract visual features from the results of stationary crowd detection. The effectiveness of the proposed algorithms is shown through experiments on a benchmark dataset.
Keywords: feature extraction; filtering theory; image motion analysis; object detection; video signal processing; video surveillance; crowd descriptors; crowd video surveillance; foreground pixel; motion filtering; quantitative analysis; spatial-temporal filtering; stationary crowd detection algorithm; stationary crowd group detection; stationary crowd groups profiling; visual feature extraction;Color;Estimation;Filtering;Indexes;Noise;Tracking;Trajectory;Stationary crowd detection; crowd video surveillance; stationary crowd analysis (ID#: 15-3599)<
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6890138&isnumber=6890121
Note:
Articles listed on these pages have been found on publicly available internet pages and are cited with links to those pages. Some of the information included herein has been reprinted with permission from the authors or data repositories. Direct any requests via Email to news@scienceofsecurity.net for removal of the links or modifications to specific citations. Please include the ID# of the specific citation in your correspondence.