Traffic Behavior Recognition Using the Pachinko Allocation Model

Abstract

CCTV-based behavior recognition systems have gained considerable attention in recent years in the transportation surveillance domain for identifying unusual patterns, such as traffic jams, accidents, dangerous driving and other abnormal behaviors. In this paper, a novel approach for traffic behavior modeling is presented for video-based road surveillance. The proposed system combines the pachinko allocation model (PAM) and support vector machine (SVM) for a hierarchical representation and identification of traffic behavior. A background subtraction technique using Gaussian mixture models (GMMs) and an object tracking mechanism based on Kalman filters are utilized to firstly construct the object trajectories. Then, the sparse features comprising the locations and directions of the moving objects are modeled by PAM into traffic topics, namely activities and behaviors. As a key innovation, PAM captures not only the correlation among the activities, but also among the behaviors based on the arbitrary directed acyclic graph (DAG). The SVM classifier is then utilized on top to train and recognize the traffic activity and behavior. The proposed model shows more flexibility and greater expressive power than the commonly-used latent Dirichlet allocation (LDA) approach, leading to a higher recognition accuracy in the behavior classification.

Development and investigation of methods of graphic-functional modeling of distributed systems

“…As noted above, all of the graph-analytic models of distributed systems, presented, in particular, in the papers [15][16][17][18][19][20][21][22], reproduce the static properties of the modeled objects. The object dynamic properties are characterized by separate models, which are integrated with the static component at the interface between the application software and the configuration files.…”

“…This issue is especially relevant to the microprocessor control systems for railway automatics, mainly for the points and signals interlocking. The technological object of such systems is the track development of railway stations, the research of graph-analytical reproduction of which is done in the papers [15][16][17][18][19][20][21][22].…”

“…Moreover, the simulation and optimization of traffic flows on the basis of a directed acyclic graph is proposed in the paper [17]. The model determines the correlation between the behavior of road users, with a built-in model of decision support.…”

“…Either way, it is necessary to generalize the methods, models and means of graph-analytical modeling of distributed technological systems and objects, set forth in the writings [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. In the aspect of the application of these categories in order to research the work and automated design of information management systems, one can distinguish such general limitations and disadvantages:…”

Hybrid Solution Combining Kalman Filtering with Takagi–Sugeno Fuzzy Inference System for Online Car-Following Model Calibration

“…Video surveillance systems prove their efficiency in road traffic patterns identification and behaviors classification. In this regard, the usage of a combination between the PAM (Pachinko allocation model) with the SVM (support vector machine) achieved better performances than the LDA (latent Dirichlet allocation) approach [ 15 ]. An innovative proposal shows that traffic patterns can be retrieved from social media applications that enable user geo-location data sharing.…”