Browsing by Author "Lin, SF"

Now showing 1 - 18 of 18

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.
An automatic focusing algorithm
(2002)
Lin, SF
;
Lin, YT
;
Su, CK
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A simple and effective automatic focusing algorithm is proposed in this article. The principle of the proposed automatic focusing algorithm is based on that, for the radial test pattern, a best-focused image should have the smallest blurred region in the middle of the acquired image, and hence, should have the smallest equivalent radius. The circular Hough transform has became a common method in numerous image-processing applications for circle detection. Various modifications to the basic circular Hough transform have been suggested, such as: the inclusion of edge orientation, simultaneous consideration of a range of circle radii, the use of a complex accumulator array with the phase proportional to the log of the radius, or for filter operations. The purpose of this work is to show that a radius of a circular region extracted by a normalized circular Hough transform is a possible solution for determining the sharpness of images. To acquire high quality images with a given CCD camera, it is crucial that the camera be located exactly at the back length of the lens, i.e., the focus position of the lens. In the best conditions, the contours of the acquired images are of the sharpest, with none of the blurring effects associated with unfocused images. Acquiring such high quality images by these means is the main goal of the automatic focusing algorithm proposed in this article. (C) 2003 Wiley Periodicals, Inc.
An incremental learning neural network for pattern classification
(1999-09-01)
Hung, CA
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A neural network architecture that incorporates a supervised mechanism into a fuzzy adaptive Hamming net (FAHN) is presented. The FAHN constructs hyper-rectangles that represent template weights in an unsupervised learning paradigm. Learning in the FAHN consists of creating and adjusting hyper-rectangles in feature space. By aggregating multiple hyper-rectangles into a single class, we can build a classifier, to be henceforth termed as a supervised fuzzy adaptive Hamming net (SFAHN), that discriminates between nonconvex and even discontinuous classes. The SFAHN can operate at a fast-learning rate in online (incremental) or offline (batch) applications, without becoming unstable. The performance of the SFAHN is tested on the Fisher iris data and on an online character recognition problem.
Design of observers with unknown inputs using eigenstructure assignment
(2000-06-01)
Lin, SF
;
Wang, AP
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Observers with unknown inputs using eigenstructure assignment are established in this paper. Orders of observers may vary from minimum order to full order. Complete and parametric solutions for observer matrices and generalized eigenvectors are obtained. Owing to the completeness and parametric forms of the solution, more properties of the observer may be obtained; hence, the solution is quite suitable for advanced applications.
Estimation of number of people in crowded scenes using perspective transformation
(2001-11-01)
Lin, SF
;
Chen, JY
;
Chao, HX
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In the past, the estimation of crowd density has become an important topic in the field of automatic surveillance systems. In this paper, the developed system goes one step further to estimate the number of people in crowded scenes in a complex background by using a single image. Therefore, more valuable information than crowd density can be obtained. There are two major steps in this system: recognition of the head-like contour and estimation of crowd size. First, the Haar wavelet transform (HWT) is used to extract the featured area of the bead-like contour, and then the support vector machine (SVM) is used to classify these featured area as the contour of a head or not. Next, the perspective transforming technique of computer vision is used to estimate crowd size more accurately. Finally, a model world is constructed to test this proposed system and the system is also applied for real-world images.
Lane detection using color-based segmentation
(2005)
Chiu, KY
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Lane boundary detection is the problem of estimating the geometric structure of the lane boundaries of a road on the images captured by a camera. To be an intelligent vehicle, lane boundary is necessary information, so the system and the algorithm should be as simple and fast as possible. In this paper, we propose a new method based on color information and this method will be applicable in complex environment. In this system, we first choose a region of interest to find out a threshold using statistical method in a color image. The threshold then will be used to distinguish possible lane boundary from the road. We use color-based segmentation to find out the lane boundary and use a quadratic function to approach it. This system demands low computational power and memory requirements, and is robust in the presence of noise, shadows, pavement, and obstacles such like cars, motorcycles and pedestrians conditions. The result images can be used as pre-processed images for lane tracking, road following or obstacle detection.
Minimizing weighted earliness and tardiness penalties in single-machine scheduling with idle time permitted
(2002-12-01)
Chen, JY
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In this paper, a single-machine scheduling problem with weighted earliness and tardiness penalties is considered. Idle time between two adjacent jobs is permitted and due dates of jobs could be unequal. The dominance rules are utilized to develop a relationship matrix, which allows a branch-and-bound algorithm to eliminate a high percentage of infeasible solutions. After combining this matrix with a branching strategy, a procedure to solve the problem is proposed. (C) 2002 Wiley Periodicals, Inc.
Preparation and epoxy curing of novel dicyclopentadiene-derived Mannich amines
(1999-03-28)
Tseng, FP
;
Chang, FC
;
Lin, SF
;
Lin, JJ
;
應用化學系
Phenol/dicyclopentadiene (DCPD) adducts were prepared from the BF3-catalyzed reaction of p-nonylphenol and dicyclopentadiene at molar ratios of 2 : 1 and 3 : 2. The phenol-terminating adducts were consequently reacted with diethylenetriamine and formaldehyde using Mannich reaction conditions. These products containing phenol, amine and tricyclodecane functionalities in the same molecule can be used as epoxy curing agents. The diethylenetriamine was add to the phenol via Mannich reaction at approximately 50% theoretical equivalent. The multiple N-H soups in amines and the O-H groups in phenols provide crosslinking sites for epoxy resins. The cured epoxy resins show improvement in tensile strength and elongation in comparison with those cured by the poly(oxypropylene) diamine (400 molecular weight) or diethylenetriamine. (C) 1999 John Wiley & Sons, Inc.
Preparation and epoxy curing of novel dicyclopentadiene-derived Mannich amines
(1997-09-07)
Lin, JJ
;
Lin, SF
;
Chang, FC
;
Tseng, FP
;
應用化學系
Preparation and epoxy curing of novel dicyclopentadiene-derived Mannich amines
(1997-09-07)
Lin, JJ
;
Lin, SF
;
Chang, FC
;
Tseng, FP
;
應用化學系
Preparation and epoxy curing of novel Mannich bases.
(1999-08-22)
Lin, JJ
;
Lin, SF
;
Tseng, FP
;
Chang, FC
;
應用化學系
Preparation and epoxy curing of p-nonylphenol/dicyclopentadiene adducts
(1999-11-28)
Tseng, FP
;
Chang, FC
;
Lin, SF
;
Lin, JJ
;
應用化學系
Phenol/dicyclopentadiene adducts were prepared from the BF3-catalyzed reaction of p-nonylphenol and dicyclopentadiene at molar ratios of 2 : 1 and 3 : 2. These dicyclopentadiene-derived novolac products contain tricyclodecane and multiple phenol functionalities. In curing with diglycidyl ether of bisphenol A, the polymer properties were compared with those cured with formaldehyde novolac or Jeffamine D-400 amine. When p-nonylphenol/dicyclopentadiene adducts were mixed with other commercially available curing agents such as Jeffamine D-400 amine, the tricyclodecane functionality was introduced into the resulting epoxy network. The flexibility of the cured resin was improved due to the presence of the tricyclodecane moiety in the polymer structure. (C) 1999 John Wiley & Sons, Inc.
Supervised adaptive hamming net for classification of multiple-valued patterns
(1997-04-01)
Hung, CA
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A Supervised Adaptive Hamming Net (SAHN) is introduced for incremental learning of recognition categories in response to arbitrary sequences of multiple-valued or binary-valued input patterns. The binary-valued SAHN derived from the Adaptive Hamming Net (AHN) is functionally equivalent to a simplified ARTMAP, which is specifically designed to establish many-to-one mappings. The generalization to learning multiple-valued input patterns is achieved by incorporating multiple-valued logic into the AHN. In this paper, we examine some useful properties of learning in a P-valued SAHN. In particular, an upper bound is derived on the number of epochs required by the P-valued SAHN to learn a list of input-output pairs that is repeatedly presented to the architecture. Furthermore, we connect the P-valued SAHN with the binary-valued SAHN via the thermometer code.
Synthesis and epoxy curing of Mannich bases derived from bisphenol A and poly(oxyalkylene)diamine
(2000-10-17)
Lin, JJ
;
Lin, SF
;
Kuo, TT
;
Chang, FC
;
Tseng, FP
A family of Mannich bases were prepared from the reaction of 2,2-bis-(4-hydroxyphenyl)propane (bisphenol A or BPA), formaldehyde, and poly(oxyalkylene)diamines at 1 : 1 : 1 or 1 : 2 : 2 molar ratio. By varying the molar ratio of bisphenol A to amine and the chemical structures of poly(oxyalkylene)diamines, a series of products with multiple functionalities of primary/secondary amines, phenols, and poly(oxyalkylene) were prepared. The curing profiles of these products toward the diglycidyl ether of bisphenol A (DGEBA) were examined by a differential scanning calorimeter (DSC). The physical properties of these cured materials were correlated with the chemical structures of the Mannich bases. Compared with the poly(oxyalkylene)diamines, the built-in phenol moiety in Mannich bases accelerated the curing rate. Both amine and phenol functionalities could be reactive sites toward diglycidyl ethers in a step-wise fashion under catalytic (triphenylphosphine) and different temperature conditions. Furthermore, the cured polymers demonstrated improved properties including tensile and flexural strength in comparison with those cured by the corresponding poly(oxyalkylene)diamines. (C) 2000 John Wiley & Sons, Inc.
The parametric solutions of eigenstructure assignment for controllable and uncontrollable singular systems
(2000-08-15)
Wang, AP
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
The previous results about the parametric solutions of eigenstructure assignment for singular systems applied by pure proportional or proportional-plus-derivative state feedback can only apply to controllable systems. It is because the parametric solutions of the right generalized eigenvectors corresponding to the infinite and uncontrollable finite eigenvalues have still not been found. In this paper, the parametric solutions of the right generalized eigenvectors for finite controllable (or uncontrollable) eigenvalues and infinite controllable (or uncontrollable) eigenvalues, when the system is applied by pure proportional or proportional-plus-derivative state feedback, are given. Hence, the parametric solution of eigenstructure assignment can be used to design the state feedback of both the controllable and uncontrollable systems. The condition for detecting the regularity of the resulting system is also given. (C) 2000 Academic Press.
The shifting of movable eigenvalues in uncontrollable singular systems
(1998-05-01)
Wang, AP
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A feedback design method for an uncontrollable singular system is proposed by which a set of possible state feedback gains can be designed to shift certain eigenvalues to other real or complex scalars. The method can be applied to single-or multi-input systems, and if so the method obtains non-unique possible feedback gains. If the original system is regular, real and impulse free, the closed-loop system also preserves these properties. (C) 1997 The Franklin Institute. Published by Elsevier Science Ltd.
Unknown input observers for singular systems designed by eigenstructure assignment
(2003-01-01)
Lin, SF
;
Wang, AP
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In this paper, any unknown input observers with orders between minimum and full orders can be established for singular systems by eigenstructure assignment method. The complete and parametric solutions for the observer matrices and for the generalized eigenvectors are obtained. The completeness and parametric forms of the solutions and the flexibility in selecting the observer order allow the designer to choose a suitable observer according to the control purposes; hence, the solutions are quite suitable for advanced applications. (C) 2003 The Franklin Institute. Published by Elsevier Science Ltd. All rights reserved.
Wavelet tree classification and hybrid coding for image compression
(2005-12-01)
Su, CK
;
Hsin, HC
;
Lin, SF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
A hybrid coding system that uses a combination of set partition in hierarchical trees (SPIHT) and vector quantisation (VQ) for image compression is presented. Here, the wavelet coefficients of the input image are rearranged to form the wavelet trees that are composed of the corresponding wavelet coefficients from all the subbands of the same orientation. A simple tree classifier has been proposed to group wavelet trees into two classes based on the amplitude distribution. Each class of wavelet trees is encoded using an appropriate procedure, specifically either SPIHT or VQ. Experimental results show that advantages obtained by combining the superior coding performance of VQ and efficient cross-subband prediction of SPIHT are appreciable for the compression task, especially for natural images with large portions of textures. For example, the proposed hybrid coding outperforms SPIHT by 0.38 dB in PSNR at 0.5 bpp for the Bridge image, and by 0.74 dB at 0.5 bpp for the Mandrill image.