學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

🏛️國立陽明交通大學機構典藏

Search Results

Now showing 1 - 10 of 12

An intelligent supervisory system for ion implantation in IC fabrication processes
(1999-02-01)
Shen, LC
;
Hsu, PL
;
電控工程研究所
;
Institute of Electrical and Control Engineering
This pager presents a real-time intelligent supervision system for IC ion implantation processes. A hardware interface is developed, to extract the features directly from the 2-D analog image signals of a beam map. A qualitative model for the beam scanning is then obtained, and the symptoms of abnormal operations can be analyzed to achieve on-line diagnosis. Furthermore, a fuzzy expert system is developed to advise operators on making appropriate adjustments for the beam scanning. This supervisory system has been implemented on Eaten NV-6200 A/AV ion implanters at the Taiwan Semiconductor Manufacturing Company. (C) 1999 Elsevier Science Ltd. All rights reserved.
A new LDA-based face recognition system which can solve the small sample size problem
(2000-10-01)
Chen, LF
;
Liao, HYM
;
Ko, MT
;
Lin, JC
;
Yu, GJ
A new LDA-based face recognition system is presented in this papal. Linear discriminant analysis (LDA) is one of the most popular linear projection techniques for feature extraction. The major drawback of applying LDA is that it may encounter the small sample size problem. In this paper, we propose a new LDA-based technique which can solve the small sample size problem. We also prove that the most expressive vectors derived ill the null space of the within-class scatter matrix using principal component analysis (PCA) are equal to the optimal discriminant vectors derived in the: original space using LDA. The experimental results show that the new LDA process improves the performance of a face recognition system significantly. (C) 2000 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.
A study on Chinese carbon-signature recognition
(2002-03-01)
Lin, SF
;
Chang, YW
;
Su, CK
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A novel off-line algorithm and a modified chain code for recognizing signatures are proposed in the present study. Carbon paper is used to detect force distributions when people write their signatures. First of all, the signature contours are located and the upper and lower profiles are generated then, these are used to classify the given si.-nature. Both the gray-scale and the chain code characteristics of a signature are used to extract structure and force distribution features, which are then transformed into a normalized vector. Finally, a Supervised Fuzzy Adaptive Hamming Network (SFAHN) is employed to interpret the feature vector in order to determine whether the signature is genuine or not. Simulation results show that the proposed algorithm has a good recognition rate.
A feature-based robust digital image watermarking scheme
(2003-04-01)
Tang, CW
;
Hang, HM
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
A robust digital image watermarking scheme that combines image feature extraction and image normalization is proposed. The goal is to resist both geometric distortion and signal processing attacks. We adopt a feature extraction method called Mexican Hat wavelet scale interaction. The extracted feature points can survive a variety of attacks and be used as reference points for both watermark embedding and detection. The normalized image of an image (object) is nearly invariant with respect to rotations. As a result, the watermark detection task can be much simplified when it is applied to the normalized image. However, because image normalization is sensitive to image local variation, we apply image normalization to nonoverlapped image disks separately. The disks are centered at the extracted feature points. Several copies of a 16-bit watermark sequence are embedded in the original image to improve the robustness of watermarks. Simulation results show that our scheme can survive low-quality JPEG compression, color reduction, sharpening, Gaussian filtering, median filtering, row or column removal, shearing, rotation, local warping, cropping, and linear geometric transformations.
Hierarchical learning architecture with automatic feature selection for multiclass protein fold classification
(2003-12-01)
Huang, CD
;
Lin, CT
;
Pal, NR
;
電控工程研究所
;
Institute of Electrical and Control Engineering
The structure classification of proteins plays a very important role in bioinformatics, since the relationships and characteristics among those known proteins can be exploited to predict the structure of new proteins. The success of a classification system depends heavily on two things: the tools being used and the features considered. For the bioinformatics applications, the role of appropriate features has not been paid adequate importance. In this investigation we use three novel ideas for multiclass protein fold classification. First, we use the gating neural network, where each input node is associated with a gate. This network can select important features in an online manner when the learning goes on. At the beginning of the training, all gates are almost closed, i.e., no feature is allowed to enter the network. Through the training, gates corresponding to good features are completely opened while gates corresponding to bad features are closed more tightly, and some gates may be partially open. The second novel idea is to use a hierarchical learning architecture (HLA). The classifier in the first level of HLA classifies the protein features into four major classes: all alpha, all beta, alpha + beta, and alpha/beta. And in the next level we have another set of classifiers, which further classifies the protein features into 27 folds. The third novel idea is to induce the indirect coding features from the amino-acid composition sequence of proteins based on the N-gram concept. This provides us with more representative and discriminative new local features of protein sequences for multiclass protein fold classification. The proposed HLA with new indirect coding features increases the protein fold classification accuracy by about 12%. Moreover, the gating neural network is found to reduce the number of features drastically. Using only half of the original features selected by the gating neural network can reach comparable test accuracy as that using all the original features. The gating mechanism also helps us to get a better insight into the folding process of proteins. For example, tracking the evolution of different gates we can find which characteristics (features) of the data are more important for the folding process. And, of course, it also reduces the computation time.
Tracking a maneuvering target using neural fuzzy network
(2004-02-01)
Duh, FB
;
Lin, CT
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A fast target maneuver detecting and highly accurate tracking technique using a neural fuzzy network based on Kalman filter is proposed in this paper. In the automatic target tracking system, there exists an important and difficult problem: how to detect the target maneuvers and fast response to avoid misstracking? The traditional maneuver detection algorithms, such as variable dimension filter (VDF) and input estimation (IE) etc., are computation intensive and difficult to implement in real time. To solve this problem, neural network algorithms have been issued recently. However, the normal neural networks such as backpropagation networks usually produce the extra problems of low convergence speed and/or large network size. Furthermore, the way to decide the network structure is heuristic. To overcome these defects and to make use of neural learning ability, a developed standard Kalman filter with a self-constructing neural fuzzy inference network (KF-SONFIN) algorithm for target tracking is presented in this paper. By generating possible target trajectories including maneuver information to train the SONFIN, the trained SONFIN can detect when the maneuver occurred, the magnitude of maneuver values and when the maneuver disappeared. Without having to change the structure of Kalman filter nor modeling the maneuvering target, this new algorithm, SONFIN, can always find itself an economic network size with a fast learning process. Simulation results show that the KF-SONFIN is superior to the traditional IE and VDF methods in estimation accuracy.
A 48.6-to-105.2 mu W Machine Learning Assisted Cardiac Sensor SoC for Mobile Healthcare Applications
(2014-04-01)
Hsu, Shu-Yu
;
Ho, Yingchieh
;
Chang, Po-Yao
;
Su, Chauchin
;
Lee, Chen-Yi
A machine- learning ( ML) assisted cardiac sensor SoC ( CS- SoC) is designed for mobile healthcare applications. The heterogeneous architecture realizes the cardiac signal acquisition, filtering with versatile feature extractions and classifications, and enables the higher order analysis over traditional DSPs. Besides, the asynchronous architecture with dynamic standby controller further suppresses the system active duty and the leakage power dissipation. The proposed chip is fabricated in a 90- nm standard CMOS technology and operates at 0.5 V- 1.0 V ( 0.7 V- 1.0 V for SRAM and I/ O interface). Examined with healthcare monitoring applications, the CS- SoC dissipates 48.6/ 105.2 mu W for real- time syndrome detections of ECG- based arrhythmia/ VCG- based myocardial infarction with 95.8/ 99% detection accuracy, respectively.
An image feature based robust digital watermarking scheme
(2002)
Tang, CW
;
Hang, HM
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
A novel robust digital image watermarking scheme which combines image feature extraction and image normalization is proposed. The goal is to resist both geometrical and signal processing attacks. We adopt a feature extraction method called Mexican Hat wavelet scale interaction. The extracted feature points can survive various attacks such as common signal processing, JPEG compression, and geometric distortions. Thus, these feature points can be used as reference points for both watermark embedding, and detection. The normalized image of a rotated image (object) is the same as the normalized version of the original image. As a result, the watermark detection task can be much simplified when it is done on the normalized image without referencing to the original image. However, because image normalization is sensitive to image local variation, we apply image normalization to non-overlapped image disks separately. The center of each disk is an extracted feature point. Several copies of one 16-bit watermark sequence are embedded in the original image to improve the robustness of watermarks. Simulation results show that our scheme can survive low quality JPEG compression, color reduction, sharpening, Gaussian filtering, median filtering, printing and scanning process, row or column removal, shearing rotation, scaling, local warping, cropping, and linear transformation.
Research friendly MPEG-7 software testbed
(2003)
Chang, FC
;
Hang, HM
;
Huang, HC
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
We design and implement a research friendly software platform, which aims at the flexibility and the abstraction of MPEG-7 application prototyping. We studied and. analyzed the MPEG-7 standard, including a typical scenario of using MPEG-7. In order to fulfill to needs of researches, in addition to the normative parts of MPEG-7, additional requirements are included. By examining these requirements, we propose a research friendly software platform. The architecture consists of a framework, utility units, and the descriptors. Because this system is implemented using Java, it also incorporates the features of the Java, environment, and thus it is flexible for developing new components and prototyping applications. We demonstrate the flexibility of this testbed by constructing an example program which allows users to manipulate image related descriptors.
Frequency subband processing and feature analysis of forward-looking ground-penetrating radar signals for land-mine detection
(2007-03-01)
Wang, Tsaipei
;
Keller, James M.
;
Gader, Paul D.
;
Sjahputera, Ozy
;
資訊工程學系
There has been significant amount of study on the use of ground-penetrating radar (GPR) for land-mine detection. This paper presents our analysis of GPR data collected at a U.S. Army test site using a new approach based on frequency subband processing. In this approach, from the radar data that have over 2.5 GHz of bandwidth, we compute separate radar images using the one wide (2 GHz) and four narrow (0.6 GHz) frequency subbands. The results indicate that signals for different frequency subbands; are significantly different and give very different performance in land-mine detection. In addition, we also examine a number of features extracted from the GPR data, including magnitude and local-contrast features, ratio between copolarization and cross-polarization signals, and features obtained using polarimetric decomposition. Feature selection procedures are employed to find subsets of features that improve detection performance when combined. Results of land-mine detection, including performance on blind test lanes, are presented.

學術出版;;Publications

Browse

Filters

Settings

Sort By

Results per page

🏛️國立陽明交通大學機構典藏

Search Results