• Journal of Korean Society for Quality Management
• /
• v.43 no.4
• /
• pp.521-532
• /
• 2015

Purpose: The purpose of this study is to compare two multivariate cumulative sum (MCUSUM) charts designed for spatio-temporal surveillance in terms of not only temporal detection performance but also spatial detection performance. Method: Experiments under various configurations are designed and performed to test two CUSUM charts, namely SMCUSUM and RMCUSUM. In addition to average run length(ARL), two measures of spatial identification accuracy are reported and compared. Results: The RMCUSUM chart provides higher level of spatial identification accuracy while two charts show comparable performance in terms of ARL. Conclusion: The RMCUSUM chart has more flexibility, robustness, and spatial identification accuracy when compared to those of the SMCUSUM chart. We recommend to use the RMCUSUM chart if control limit calibration is not an urgent task.

• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.612-628
• /
• 2020

This paper presents a new approach for detecting abnormal behaviors in complex surveillance scenes where anomalies are subtle and difficult to distinguish due to the intricate correlations among multiple objects' behaviors. Specifically, a cascaded probabilistic topic model was put forward for learning the spatial context of local behavior and the temporal context of global behavior in two different stages. In the first stage of topic modeling, unlike the existing approaches using either optical flows or complete trajectories, spatio-temporal correlations between the trajectory fragments in video clips were modeled by the latent Dirichlet allocation (LDA) topic model based on Markov random fields to obtain the spatial context of local behavior in each video clip. The local behavior topic categories were then obtained by exploiting the spectral clustering algorithm. Based on the construction of a dictionary through the process of local behavior topic clustering, the second phase of the LDA topic model learns the correlations of global behaviors and temporal context. In particular, an abnormal behavior recognition method was developed based on the learned spatio-temporal context of behaviors. The specific identification method adopts a top-down strategy and consists of two stages: anomaly recognition of video clip and anomalous behavior recognition within each video clip. Evaluation was performed using the validity of spatio-temporal context learning for local behavior topics and abnormal behavior recognition. Furthermore, the performance of the proposed approach in abnormal behavior recognition improved effectively and significantly in complex surveillance scenes.

• Journal of Intelligence and Information Systems
• /
• v.10 no.2
• /
• pp.15-37
• /
• 2004

With the growth in the size of datasets, data mining has recently become an important research topic. Especially, interests about spatio-temporal data mining has been increased which is a method for analyzing massive spatio-temporal data collected from a wide variety of applications like GPS data, trajectory data of surveillance system and earth geographic data. In the former approaches, conventional clustering algorithms are applied as spatio-temporal data mining techniques without any modification. In this paper, we focused to SOM that is the most common clustering algorithm applied to clustering analysis in data mining wet and develop the spatio-temporal data mining module based on it. In addition, we analyzed the clustering results of developed SOM module and compare them with those of K-means and Agglomerative Hierarchical algorithm in the aspects of homogeneity, separation, separation, silhouette width and accuracy. We also developed specialized visualization module fur more accurate interpretation of mining result.

• 대한예방의학회:학술대회논문집
• /
• 2005.10a
• /
• pp.482-482
• /
• 2005

• Journal of the Korea Institute of Military Science and Technology
• /
• v.14 no.5
• /
• pp.913-919
• /
• 2011

In this paper, we present a new efficient event detection algorithm for sensor networks with faults. We focus on multi-attributed events, which are sets of data points that correspond to interesting or unusual patterns in the underlying phenomenon that the network monitors. Conventional algorithms cannot detect some events because they treat only their own sensor readings which can be affected easily by environmental or physical problem. Our approach exploits spatio-temporal correlation of sensor readings. Sensor nodes exchange a fault-tolerant code encoded their own readings with neighbors, organize virtual sensor readings which have spatio-temporal correlation, and determine a result for multi-attributed events from them. In the result, our proposed algorithm provides improvement of detecting multi-attributed events and reduces the number of false-negatives due to negative environmental effects.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.473-478
• /
• 2015

Recently, automatic video monitoring of individual pigs is emerging as an important issue in the management of group-housed pigs. Although a rich variety of studies have been reported on video monitoring techniques in intensive pig farming, it still requires further elaboration. In particular, when there exist adjoining pigs in a crowd pig room, it is necessary to have a way of separating adjoining pigs from the perspective of an image processing technique. In this paper, we propose an efficient image segmentation solution using both spatio-temporal information and region growing method for the identification of individual pigs in video surveillance systems. The experimental results with the videos obtained from a pig farm located in Sejong illustrated the efficiency of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.1
• /
• pp.169-189
• /
• 2015

Detection of anomalous events from video streams is a challenging problem in many video surveillance applications. One such application that has received significant attention from the computer vision community is traffic video surveillance. In this paper, a Lossy Count based Sequential Temporal Pattern mining approach (LC-STP) is proposed for detecting spatio-temporal abnormal events (such as a traffic violation at junction) from sequences of video streams. The proposed approach relies mainly on spatial abstractions of each object, mining frequent temporal patterns in a sequence of video frames to form a regular temporal pattern. In order to detect each object in every frame, the input video is first pre-processed by applying Gaussian Mixture Models. After the detection of foreground objects, the tracking is carried out using block motion estimation by the three-step search method. The primitive events of the object are represented by assigning spatial and temporal symbols corresponding to their location and time information. These primitive events are analyzed to form a temporal pattern in a sequence of video frames, representing temporal relation between various object's primitive events. This is repeated for each window of sequences, and the support for temporal sequence is obtained based on LC-STP to discover regular patterns of normal events. Events deviating from these patterns are identified as anomalies. Unlike the traditional frequent item set mining methods, the proposed method generates maximal frequent patterns without candidate generation. Furthermore, experimental results show that the proposed method performs well and can detect video anomalies in real traffic video data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3991-4007
• /
• 2022

In the public healthcare, a computational system that can automatically and efficiently detect and classify falls from a video sequence has significant potential. With the advancement of deep learning, which can extract temporal and spatial information, has become more widespread. However, traditional 3D CNNs that usually adopt shallow networks cannot obtain higher recognition accuracy than deeper networks. Additionally, some experiences of neural network show that the problem of gradient explosions occurs with increasing the network layers. As a result, an enhanced three-dimensional ResNet-based method for fall detection (3D-ERes-FD) is proposed to directly extract spatio-temporal features to address these issues. In our method, a 50-layer 3D residual network is used to deepen the network for improving fall recognition accuracy. Furthermore, enhanced residual units with four convolutional layers are developed to efficiently reduce the number of parameters and increase the depth of the network. According to the experimental results, the proposed method outperformed several state-of-the-art methods.

• Journal of Broadcast Engineering
• /
• v.20 no.3
• /
• pp.449-463
• /
• 2015

Video object tracking is important for variety of applications, such as security, video indexing and retrieval, video surveillance, communication, and compression. This paper proposes an object tracking method in HEVC bitstreams. Without pixel reconstruction, motion vector (MV) and size of prediction unit in the bitstream are employed in an Spatio-Temporal Markov Random Fields (ST-MRF) model which represents the spatial and temporal aspects of the object's motion. Coefficient-based object shape adjustment is proposed to solve the over-segmentation and the error propagation problems caused in other methods. In the experimental results, the proposed method provides on average precision of 86.4%, recall of 79.8% and F-measure of 81.1%. The proposed method achieves an F-measure improvement of up to 9% for over-segmented results in the other method even though it provides only average F-measure improvement of 0.2% with respect to the other method. The total processing time is 5.4ms per frame, allowing the algorithm to be applied in real-time applications.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2012.07a
• /
• pp.298-301
• /
• 2012

This paper presents a graph-based method of detecting and tracking moving objects in H.264/SVC bitstreams for video surveillance applications that makes use the information from spatial base and enhancement layers of the bitstreams. In the base layer, segmentation of real moving objects are first performed using a spatio-temporal graph by removing false detected objects via graph pruning and graph projection, followed by graph matching to precisely identify the real moving objects over time even under occlusion. For the accurate detection and reliable tracking of moving objects in the enhancement layer, as well as saving computational complexity, the identified block groups of the real moving objects in the base layer are then mapped to the enhancement layer to provide accurate and efficient object detection and tracking in the bitstreams of higher resolution. Experimental results show the proposed method can produce reliable results with low computational complexity in both spatial layers of H.264/SVC test bitstreams.