Limits...
Traffic Behavior Recognition Using the Pachinko Allocation Model.

Huynh-The T, Banos O, Le BV, Bui DM, Yoon Y, Lee S - Sensors (Basel) (2015)

Bottom Line: 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.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Engineering, Kyung Hee University, Suwon 446-701, Korea. thienht@oslab.khu.ac.kr.

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 PAMinto 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.

No MeSH data available.


Related in: MedlinePlus

Illustration of SVM-binary tree architecture (BTA).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541867&req=5

f4-sensors-15-16040: Illustration of SVM-binary tree architecture (BTA).

Mentions: Based on the PAM-based topic modeling, every video sequence can be represented through a u × v matrix, where u is the number of behaviors and v is the number of activities. To train the classifier, the labels of vectors and matrices are manually denoted with their classes manually In this paper, the authors use a SVM with binary tree architecture (SVM-BTA) [38] to solve the N-class pattern recognition problem. An illustration of SVM-BTA is shown in Figure 4. Each node in the architecture makes a binary decision using the original SVM. By recursively dividing the classes into two disjointed groups in each node of the decision tree, the SVM classifier decides the group to which the unknown samples that should be assigned. The class is determined by a clustering algorithm according to the class membership and the inter-class distance. Although N − 1 SVMs are trained for an N-class problem, only log2N SVMs are consulted at most to classify a sample. This approach requires fewer binary SVMs than popular methods, such as N (N − 1)/2 SVMs in the one-against-one approach and N SVMs in the one-against-others approach. Moreover, both approaches have the drawback of very expensive computational cost requirements and accuracy degradation [38]. An essential contribution of the SVM-BTA approach, the multiclass issue, is converted into binary-tree architectures without performance reduction. Moreover, a dramatic improvement in recognition speed can be achieved for increasing the number of classes.


Traffic Behavior Recognition Using the Pachinko Allocation Model.

Huynh-The T, Banos O, Le BV, Bui DM, Yoon Y, Lee S - Sensors (Basel) (2015)

Illustration of SVM-binary tree architecture (BTA).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541867&req=5

f4-sensors-15-16040: Illustration of SVM-binary tree architecture (BTA).
Mentions: Based on the PAM-based topic modeling, every video sequence can be represented through a u × v matrix, where u is the number of behaviors and v is the number of activities. To train the classifier, the labels of vectors and matrices are manually denoted with their classes manually In this paper, the authors use a SVM with binary tree architecture (SVM-BTA) [38] to solve the N-class pattern recognition problem. An illustration of SVM-BTA is shown in Figure 4. Each node in the architecture makes a binary decision using the original SVM. By recursively dividing the classes into two disjointed groups in each node of the decision tree, the SVM classifier decides the group to which the unknown samples that should be assigned. The class is determined by a clustering algorithm according to the class membership and the inter-class distance. Although N − 1 SVMs are trained for an N-class problem, only log2N SVMs are consulted at most to classify a sample. This approach requires fewer binary SVMs than popular methods, such as N (N − 1)/2 SVMs in the one-against-one approach and N SVMs in the one-against-others approach. Moreover, both approaches have the drawback of very expensive computational cost requirements and accuracy degradation [38]. An essential contribution of the SVM-BTA approach, the multiclass issue, is converted into binary-tree architectures without performance reduction. Moreover, a dramatic improvement in recognition speed can be achieved for increasing the number of classes.

Bottom Line: 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.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Engineering, Kyung Hee University, Suwon 446-701, Korea. thienht@oslab.khu.ac.kr.

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 PAMinto 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.

No MeSH data available.


Related in: MedlinePlus