Browsing by Author "Tseng, CL"

Now showing 1 - 7 of 7

A Self-growing Probabilistic decision-based neural network for anchor/speaker identification
(2003)
Chen, YH
;
Tseng, CL
;
Fu, HC
;
Pao, HT
;
資訊工程學系
In this paper, we propose a new clustering algorithm for a mixture Gaussian based neural network, called Self-growing Probabilistic decision-based neural networks (SPDNN). The proposed Self-growing cluster learning (SGCL) algorithm is able to find the natural number of prototypes based on a self-growing validity measure, Bayesian Information Criterion (BIC). The learning process starts with a single prototype randomly initialized in the feature space and grows adaptively during the learning process until most appropriate number of prototypes are found. We have conduct numerical and real world experiments to demostrate the effectiveness of the SGCL algorithm. In the results of using SGCL to trainin the SPDNN for anchor/speaker identification, we have observed noticeable improvement among various model-based or vector quantization-based classification schemes.
A self-growing probabilistic decision-based neural network with automatic data clustering
(2004-10-01)
Tseng, CL
;
Chen, YH
;
Xu, YY
;
Pao, HT
;
Fu, HC
In this paper, we propose a new clustering algorithm for a mixture of Gaussian-based neural network and self-growing probabilistic decision-based neural networks (SPDNN). The proposed self-growing cluster learning (SGCL) algorithm is able to find the natural number of prototypes based on a self-growing validity measure, Bayesian information criterion (BIC). The learning process starts from a single prototype randomly initialized in the feature space and grows adaptively during the learning process until most appropriate number of prototypes are found. We have conducted numerical and real-world experiments to demonstrate the effectiveness of the SGCL algorithm. In the results of using SGCL to train the SPDNN for data clustering and speaker identification problems, we have observed a noticeable improvement among various model-based or vector quantization-based classification schemes. (C) 2004 Elsevier B.V. All rights reserved.
Automated information. mining on multimedia TV news archives
(2005)
Lai, PS
;
Cheng, SS
;
Sun, SY
;
Huang, TY
;
Su, JM
This paper addresses an integrated information mining techniques for multimedia TV-news archive. The utilizes techniques from the fields of acoustic, image, and video analysis, for information retrieval on news story title, newsman and scene identification. The goal is to construct a compact yet meaningful abstraction of broadcast news video, allowing users to browse through large amounts of data in a non-linear fashion with flexibility and efficiency. By using acoustic analysis, the system can classify video into news versus commercials, with 90% accuracy on a data set of 400 hours TV-news recorded off the air from July 2003 to August of 2004. By applying speaker identification and/or image detection techniques, each news stories can be segmented with an accuracy of 96%. On screen captions or subtitles are recognized by OCR techniques to produce the text title of each news stories. The extracted title words can be used to link or to navigate more related News contents on the WWW. In cooperation with facial and scene analysis and recognition techniques, OCR results can provide users with multimodality query for specific news stories. Some experimental results are presented and discussed for the system reliability and performance evaluation and comparison.
Characterization and modeling of the metal diffusion from deep ultraviolet photoresist and silicon-based substrate
(2001-05-01)
Wang, TK
;
Wang, MY
;
Ko, FH
;
Tseng, CL
;
奈米中心
The radioactive tracer technique was applied to investigate the out-diffusion of the transition metals (Cu, Fe and Co) from deep ultraviolet (DUV) photoresist into underlying substrate. Two important process parameters, viz., baking temperatures and substrate types (i.e., bare silicon, polysilicon, silicon oxide and silicon nitride), were evaluated. Results indicate that the out-diffusion of Co is insignificant, irrespective of the substrate type and baking temperature. The Out diffusion of Cu is significant for substrates of bare silicon and polysilicon but not for silicon oxide and nitride; for Fe, the story is reversed. The substrate type appears to strongly affect the diffusion, while the baking temperature does not. Also, the effect of solvent evaporation was found to play an important role in impurity diffusion. Using the method of numerical analysis, a diffusion profile was depicted in this work to describe the out-diffusion of metallic impurities from photoresist layer under Various baking conditions. In addition, the effectiveness of various wet-cleaning recipes in removing metallic impurities such as Cu, Fe and Co was also studied using the radioactive tracer technique. Among the six cleaning solutions studied, SC2 and SPM are the most effective in impurity removal. An out-diffusion cleaning model was first proposed to describe the cleaning process. A new cleaning coefficient, h(T), was suggested to explain the cleaning effect. The cleaning model could explain the tracer results. (C) 2001 Elsevier Science Ltd. All rights reserved.
Cluster-based support vector machines in text-independent speaker identification
(2004)
Sun, SY
;
Tseng, CL
;
Chen, YH
;
Chuang, SC
;
Fu, HC
Based on Statistical learning theory, Support Vector Machines(SVM) is a powerful tool for various classification problems, such as pattern recognition and speaker identification etc. However, Training SVM consumes large memory and long computing time. This paper proposes a cluster-based learning methodology to reduce training time and the memory size for SVM. By using k-means based clustering technique, training data at boundary of each cluster were selected for SVM learning. We also applied this technique to text-independent speaker identification problems. Without deteriorating recognition performance, the training data and time can be reduced up to 75% and 87.5% respectively.
COSMIC: Geodetic applications in improving earth's gravity model
(2000-03-01)
Chao, BF
;
Pavlis, EC
;
Hwang, CW
;
Liu, CC
;
Shum, CK
The routine GPS tracking data that will be collected for the precise orbit determination of the eight COSMIC spacecraft will contain valuable information about the terrestrial gravity field. Numerical simulations based on current straw-man mission scenarios have shown: (1) For the static model of the gravity field, significant improvement is possible over our present knowledge embodied in the EGM96 model, out to harmonic degree as high as 40. (2) With respect to time-varying gravitational signals, large-scale mass transport processes in the geophysical fluids (e.g., atmosphere and oceans) can be detected or monitored during the mission, providing important global change information. Further sophisticated simulations incorporating more realistic force models will be needed once COSMIC mission scenario and spacecraft design are completely defined.
Multimedia TV news browsing system
(2004)
Xu, YY
;
Chen, YH
;
Tseng, CL
;
Lai, PS
;
Fu, HC
In this demo, we present a system we have developed for automatic broadcast-quality video indexing that successfully combines results from the fields of speaker verification, acoustic analysis, very large vocabulary caption character recognition, content based sampling of video, information retrieval, dialogue systems, and ASF media delivery over IP. The prototype system of this demo is available at http://nn.csie.nctu.edu.tw/Project1-1/intruduction.htm.