A People-Counting System Using a Hybrid RBF Neural Network.

Neural Processing Letters (Impact Factor: 1.24). 01/2003; 18:97-113. DOI: 10.1023/A:1026226617974
Source: DBLP

ABSTRACT A people-counting system using hybrid RBF neural network is described. The proposed system is effective and flexible for the purpose of performing on-line people counting. Compared with other conventional approach, this system introduces a novel method for feature extraction. In this Letter, a new type of hybrid RBF network is developed to enhance the classification performance. The hybrid RBF based people-counting system is thoroughly compared with other approaches. Extensive and promising results were obtained and the analysis indicates that the proposed hybrid RBF based system provides excellent people-counting results in an open passage. A supervised clustering method is proposed for initialising the hybrid RBF network. In order to substantiate the introduction of the hybrid RBF and the proposed supervised clustering algorithm, test results on a vowel recognition benchmark dataset are also included in the Letter.

  • [Show abstract] [Hide abstract]
    ABSTRACT: As a particular class of public security issues, the large-scale crowd analysis plays a very important role in video surveillance application. This paper proposes a sparse spatial-temporal local binary pattern (SST-LBP) descriptor to extract dynamic texture of the walking crowd which can be applied to the crowd density estimation and distribution analysis. The proposed approach consists of four steps. First of all, sparse selected locations are extracted, which vary notably in both spatial domain and temporal domain. Afterwards, we propose a SST-LBP algorithm to extract the local dynamic feature and utilize the local feature’s statistical property to describe the crowd feature. Thirdly, the overall crowd density level can be determined by classifying the crowd feature with support vector machine. Finally, the local feature is used to represent the local density and then the overall density distribution can be described. To improve the accuracy, we introduce the perspective correction into the detection of sparse selected locations and the spectrum analysis of SST-LBP code. The experiments on different datasets not only show that the proposed SST-LBP method is effective and robust on the large-scale crowd density estimation and distribution, but also indicate that the deformity correction is useful. Compared with other methods, the proposed method has the advantage of low computation complexity and high efficiency. In addition, it performs well on all density levels and can present local crowd distribution.
    Multimedia Tools and Applications 11/2014; 73(1). DOI:10.1007/s11042-012-1264-0 · 1.06 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Automatically counting vehicles in complex traffic scenes from videos is challenging. Detection and tracking algorithms may fail due to occlusions, scene clutters, and large variations of viewpoints and vehicle types. We propose a new approach of counting vehicles through exploiting contextual regularities from scene structures. It breaks the problem into simpler problems, which count vehicles on each path separately. The model of each path and its source and sink add strong regularization on the motion and the sizes of vehicles and can thus significantly improve the accuracy of vehicle counting. Our approach is based on tracking and clustering feature points and can be summarized in threefold. First, an algorithm is proposed to automatically learn the models of scene structures. A traffic scene is segmented into local semantic regions by exploiting the temporal cooccurrence of local motions. Local semantic regions are connected into global complete paths using the proposed fast marching algorithm. Sources and sinks are estimated from the models of semantic regions. Second, an algorithm is proposed to cluster trajectories of feature points into objects and to estimate average vehicle sizes at different locations from initial clustering results. Third, trajectories of features points are often fragmented due to occlusions. By integrating the spatiotemporal features of trajectory clusters with contextual models of paths and sources and sinks, trajectory clusters are assigned into different paths and connected into complete trajectories. Experimental results on a complex traffic scene show the effectiveness of our approach.
    IEEE Transactions on Intelligent Transportation Systems 06/2013; 14(2):1016-1022. DOI:10.1109/TITS.2013.2248001 · 2.47 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes a vision based pedestrian detection and tracking system which is able to count people in very crowded situations like escalator entrances in underground stations. The proposed system uses motion to compute regions of interest and prediction of movements, extracts shape information from the video frames to detect individuals, and applies texture features to recognize people. A search strategy creates trajectories and new pedestrian hypotheses and then filters and combines those into accurate counting events. We show that counting accuracies up to 98 % can be achieved.
    Video and Signal Based Surveillance, 2006. AVSS '06. IEEE International Conference on; 12/2006