http://hdl.handle.net/10453/86 2013-05-24T04:13:49Z http://hdl.handle.net/10453/18296 Content - aware broadcast soccer video retargeting using fuzzy logic Gao L; Xu Min; Yan S.; Liu Mg; Hou Ch A content-aware video retargeting method is proposed for playing broadcast soccer video in small displays. Four visual perception clues are predefined based on soccer game-specific knowledge and modelled by visual attention features firstly. Then, a fuzzy logic inference system is proposed to estimate visual attention values (AVs) of ball and players by fusing attention features. AVs are later used to determine the region of interest (ROI) of each frame. Finally, a retargeted video is generated by the ROI of each frame with polynomial curve fitting for temporal smoothing. Both subjective and objective evaluation results are promising. 2011-01-01T00:00:00Z http://hdl.handle.net/10453/18297 Image Classification Technique using Modified Particle Swarm Optimization Shukran Mohd; Chung Yuk; Yeh Wei-Chang; Wahid Noorhaniza; Zaidi Ahmad Image classification is becoming ever more important as the amount of available multimedia data increases. With the rapid growth in the number of images, there is an increasing demand for effective and efficient image indexing mechanisms. For large image databases, successful image indexing will greatly improve the efficiency of content based image classification. One attempt to solve the image indexing problem is using image classification to get high-level concepts. In such systems, an image is usually represented by various low-level features, and high-level concepts are learned from these features. PSO has recently attracted growing research interest due to its ability to learn with small samples and to optimize high-dimensional data. Therefore, this paper will introduce the related work on image feature extraction. Then, several techniques of image feature extraction will be introduced which include two main methods. These methods are RGB and Discrete Cosine Transformation (DCT). Finally, several experimental designs and results concerning the application of the proposed image classification using modified PSO classifier will be described in detail. 2011-01-01T00:00:00Z http://hdl.handle.net/10453/18299 Approximate Reliability Function Based on Wavelet Latin Hypercube Sampling and Bee Recurrent Neural Network Yeh Wei-Chang; Hsieh Tsung-Jung; Chih Mingchang; Liu Sin-Long; Ping Su This work combines a Bee Recurrent Neural Network (BRNN) optimized by the Artificial Bee Colony (ABC) algorithm with Monte Carlo Simulation (MCS) to generate a novel approximate model for predicting network reliability. We utilize the Wavelet Transform (WT)-based Latin Hypercube Sampling (LHS) (WLHS) to select input training data, and open the black box of neural networks by constructing a limited space reliability function from neural network parameters. Furthermore, the proposed method compares favorably with existing methods in literature based on experimental results for a benchmark example. The result reveals that the novel WLHS-MCS based on BRNN (WLHS-BRNN-MCS for short) is an excellent estimator of the reliability function. 2011-01-01T00:00:00Z http://hdl.handle.net/10453/18298 A Sequential Decomposition Method for Estimating Flow in a Multi-Commodity, Multistate Network Yeh Wei-Chang The weighted multi-commodity multistate unreliable network (WMMUN) is an extension of the multistate network. It is a new network composed of multistate unreliable components (such as arcs and nodes) with various weight capacities which is able to transmit different types of commodities. Currently, the method used to calculate the direct WMMUN reliability is derived from algorithms based on D-minimal path (D-MP). The best-known method may fail to find real D-MPs, and therefore requires more comparison and verification. A very simple algorithm based on the sequential decomposition method has been developed for finding all real D-MPs before calculating the WMMUN reliability. The relationships among the different versions of WMMUN reliability problems have also been clarified. The correctness and computational complexity of the proposed algorithm will be analysed and proven in this paper. An example will be given to illustrate how the WMMUN reliability is evaluated using the proposed algorithm. Computational results compare favorably with existing methods in terms of the running time, the number of d-MPs, and the number of D-MP candidates. 2011-01-01T00:00:00Z