Browsing by Author "Hsu, HH"

Now showing 1 - 15 of 15

Adaptive self-quantization in wavelet-based fractal image compression
(1999-05-01)
Wu, BF
;
Hu, YQ
;
Hsu, HH
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Finding a model to quantize the scale factors in wavelet-based fractal image compression is a complicated issue. To avoid error, it is helpful to model the distribution of the scale factors and quantize them before the computation process by iterated function systems. Traditionally, a fixed model with uniform distribution was frequently adopted. This is not sophisticated enough, however, to quantize these scale factors from errors since, in general, the factors are not uniformly distributed. We propose an adaptive algorithm with self-quantization to overcome this drawback. Except for the functions of adaptation and self-quantification, the approach has the optimal property that the fundamental objective is to reduce the quantization errors.
An innovation approach to optimize a Kalman filter with an H-infinity error bound by secant method
(1996-07-01)
Wu, BF
;
Hsu, HH
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
Designing a Kalman filter with a constraint on the H-infinity norm of the estimation error was first developed by Bernstein and Haddad in 1989. The main result is a sufficient condition for characterizing the Kalman filter. In this paper, similar to the standard Kalman filter, the properties of orthogonal principles are also shown to be preserved. Furthermore, the uniqueness, as opposed to an H-infinity filter, of the filter is implied by the orthogonal principles. An innovative approach to obtaining the minimum energy with a constraint on the H-infinity norm of the estimation en ol is proposed since the original work of Bernstein and Haddad does not, in general, reach the minimum energy of the estimation error. By means of the Secant method, the energy of the estimation error can be reduced as much as possible, under the condition that the H-infinity error bound is still satisfied.
An integrated method in wavelet-based image compression
(1998-08-01)
Hsu, HH
;
Hu, YQ
;
Wu, BF
;
電控工程研究所
;
Institute of Electrical and Control Engineering
In this paper, we propose an integrated image compression method, which contains the discrete wavelet transform, scalar quantization and some lossless codings, to gain higher compression ratios while maintaining the image fidelity. The discrete wavelet transform has the properties of entropy reduction and energy concentration in high frequency subimages. An innovative approach, called revised rein-length coding, is proposed to improve the compression performance. The idea of this approach is to represent the appearance of symbols of the run-length cone in exponential expression for saving the storage in bits. The differential pulse code modulation is used to reduce the entropy of the lowest frequency subimage of the discrete wavelet transform and to achieve high compression effects losslessly. (C) 1998 The Franklin Institute. Published by Elsevier Science Ltd.
An integrated method to image compression using the discrete wavelet transform
(1997)
Hu, YQ
;
Hsu, HH
;
Wu, BF
;
交大名義發表
;
電控工程研究所
In this paper, we propose an integrated image compression method, which contains the discrete wavelet tranform scalar quantization and some lossless codings, to gain higher compression ratios while maintaining the image fidelity. The discrete wavelet transform, a multiresolution technique, has the properties of entropy reduction and energy concentration in high frequency subimages. Scalar quantizer is applied to the high frequency components since those histograms can be modelled to the generalized Gaussian distribution. These subimages with small relative energy can be dropped entirely to compensate the compression ratio if the optimal scalar quantization is adopted. An innovative approach, named as revised run-length coding, is proposed to improve the compression performance. The idea of this approach is to represent the appearance of symbols of run-length codes in exponential expression for saving the storage In bits. One coding method, differential pulse coded modulation, is introduced to reduce the entropy of the lowest frequency subimage performed after the discrete wavelet transform and to acheive the high compression effect losslessly. The experimental results are compared with some well-known methods, for example, JPEG, (entropy constrained) vector quantization, fractal-based compression method and wavelet with variable-length coding.
Chamber calibration of CPT under simulated field conditions
(1998)
Huang, AB
;
Hsu, HH
;
土木工程學系
;
Department of Civil Engineering
An important drawback of calibrating CPT in a chamber is its boundary effects. The lateral boundary of the specimen in a conventional calibration chamber is either stress controlled or rigid. Neither of these two boundary conditions can correctly duplicate the field conditions which lies somewhere in between. In order to eliminate the boundary effects, a new calibration chamber system that enables the field conditions be simulated has been developed. The new chamber wall, 800mm in diameter and 1600mm high, is made of 20 vertically stacked steel rings. The boundary movement induced by the cone penetration is measured at each ring level by an extensometer. The lateral stress at each ring level is determined by the stress-strain relationship of the specimen. A series of CPT calibration tests has been performed in a uniform, dry silica sand using the new system. Results show that the new simulator is capable of eliminating boundary effects, even at a relative density of 84%. The paper describes the design of the new chamber system and presents available test data.
Cone penetration tests under simulated field conditions
(2005-06-01)
Huang, AB
;
Hsu, HH
;
土木工程學系
;
Department of Civil Engineering
Cone penetration tests (CPT) performed in a conventional cavity-wall calibration chamber are inevitably affected by boundary conditions. The authors have developed an axisymmetric field simulator whereby CPT calibration tests can be conducted under substantially reduced boundary effects. The simulator consists of a cylindrical, physical sand specimen and a stack of expandable rings to simulate the soil from beyond the physical boundary. A series of CPT calibration tests has been performed in a clean, uniformly graded quartz sand using this simulator system. For a given relative density, the cone tip resistance (q(c)) was found to be most closely related to mean normal stress. The horizontal stress at the physical boundary increased during CPT under simulated field conditions. This increased horizontal stress had a much more significant and consistent effect on q(c) than the initial horizontal stress existing in the soil prior to CPT. The paper describes the results of CPT in the simulator, and discusses implications for the performance of CPT in a chamber and their interpretation.
Development of an axisymmetric field simulator for cone penetration tests in sand
(1998-12-01)
Hsu, HH
;
Huang, AB
;
土木工程學系
;
Department of Civil Engineering
An experimental system capable of simulating axisymmetric field conditions has been developed. This system, as an improvement over conventional calibration chambers, enables cone penetration tests (CPT) to be calibrated under minimal boundary effects. The new simulator consists of a stack of 20 rings to house the sand specimen. These rings are lined with an inflatable silicone rubber membrane on the inside. The boundary expansion and stress are measured and individually controlled, respectively, at each ring level during the CPT. The soil from physical boundary to infinity is simulated using a nonlinear cavity expansion curve derived from a lateral compression test on the specimen. Results from a series of CPTs show that at relative densities of approximately 80%, the cone tip resistance values agree within 4% as the diameter ratio of the physical specimen over cone varies from 18 to 22. This paper describes unique features of this field simulator, presents available CPT data performed in the simulator, and discusses its implications on future calibration tests.
Entropy-constrained scalar quantization and minimum entropy with error bound by discrete wavelet transforms in image compression
(2000-04-01)
Wu, BF
;
Hsu, HH
;
電控工程研究所
;
Institute of Electrical and Control Engineering
The global maximum of an entropy function with different decision levels for a three-level scaler quantizer performed after a discrete wavelet transform was derived. Herein, we considered the case of entropy-constrained scalar quantization capable of avoiding many compression ratio reductions as the mean squared error was minimized. We also dealt with the problem of minimum entropy with an error bound, which was referred to as the rate distortion function, For generalized Gaussian distributed input signals, the Shannon bound would decrease monotonically when the parameter of distribution gamma was to leave from 2. That is, the Gaussian distributions would contain the highest Shannon bound among the generalized Gaussian distributions. Additionally, we proposed two numerical approaches of the secant and false position methods implemented in real cases to solve the problems of entropy-constrained scalar quantization and minimum entropy with an error bound, The convergence condition of the secant method was also addressed.
Knowledge mining from trained neural networks
(2002-06-01)
Su, CT
;
Hsu, HH
;
Tsai, CH
;
交大名義發表
;
National Chiao Tung University
Despite their diverse applications in many domains, neural networks are difficult to interpret owning the lack of mathematical models to express the training result. While adopting the rule extraction method to develop different algorithms, many researchers normally simplify a network's structure and then extract rules from the simplified networks, This type of data limits such conventional approaches when attempting to remove the unnecessary connections. In addition to developing network pruning and extraction algorithms, this work attempts to determine the important input nodes. In the proposed algorithms, the type of input data is not limited to binary, discrete or continuous. Moreover, two numerical examples are analyzed. Comparing the results from the proposed algorithms with those from See5 demonstrates the effectiveness of the proposed algorithms.
Local displacement measurement and physical modeling
(2003)
Huang, AB
;
Wu, CT
;
Hsu, HH
;
土木工程學系
;
Department of Civil Engineering
The importance and value of local displacement measurement in physical modeling and/or field tests have been well recognized. This paper describes two cases where local displacement was monitored for cone penetration tests (CPT) in a calibration chamber and model foundation load tests. In the CPT chamber calibration tests, local displacement measurement was used as part of a scheme to minimize boundary effects. In the model foundation load tests, the local displacement measurement was used to monitor the distribution of displacement in the rock mass. This paper describes the techniques implemented in these two cases and discuses their performances.
Minimization of the estimation error of the Kalman filter with an H-infinity error bound by secant method
(1995)
Wu, BF
;
Hsu, HH
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Numerical solution for the second-order wave interaction with porous structures
(1999-02-15)
Hsu, HH
;
Wu, YC
;
土木工程學系
;
Department of Civil Engineering
This study mathematically formulates the fluid field of a water-wave interaction with a porous structure as a two-dimensional, non-linear boundary value problem (bvp) in terms of a generalized velocity potential. The non-linear bvp is reformulated into an infinite set of linear bvps of ascending order by Stokes perturbation technique, with wave steepness as the perturbation parameter. Only the first- and second-order linear bvps are retained in this study. Each linear bvp is transformed into a boundary integral equation. In addition, the boundary element method (BEM) with linear elements is developed and applied to solve the first- and second-order integral equations. The first- and second-order wave profiles, reflection and transmission coefficients, and the amplitude ratio of the second-order components are computed as well. The numerical results correlate well with previous analytical and experimental results. Numerical results demonstrate that the second-order component can be neglected for a deep water-wave and may become significant for an intermediate depth wave. Copyright (C) 1999 John Wiley & Sons, Ltd.
Scattering of water wave by a submerged horizontal plate and a submerged permeable breakwater
(1999-04-01)
Hsu, HH
;
Wu, YC
;
土木工程學系
;
Department of Civil Engineering
Based on a two-dimensional linear water wave theory, this study develops the boundary element method (BEM) to examine normally incident wave scattering by a fixed, submerged, horizontal, impermeable plate and a submerged permeable breakwater in water of finite depth. Numerical results for the transmission coefficients are also presented. In addition, the numerical technique's accuracy is demonstrated by comparing the numerical results with previously published numerical and experimental ones. According to that comparison, the transmission coefficient relies not only on the submergence of the horizontal impermeable plate and the height of the permeable breakwater, but also on the distance between horizontal plate and permeable breakwater. Results presented herein confirm that the transmission coefficient is minimum for the distance approximately equal to four times the water depth. (C) 1998 Elsevier Science Ltd. All rights reserved.
The behavior of a compressible silty fine sand
(1999-02-01)
Huang, AB
;
Hsu, HH
;
Chang, JW
;
土木工程學系
;
Department of Civil Engineering
Publications associated with sands are often limited to clean (i.e., little fines content), uniform, uncemented silica or quartz sand. On the other hand, the importance of mineral content, particle shapes, as well as gradation to the behavior of sand has long been recognized. Although systematic studies of sands ether than clean quartz sand have been limited, there is increasing attention being paid to sands with an appreciable fines content. Because of a major construction project, extensive field and laboratory experiments were performed on a silty fine sand from Mai-Liao, which is located on the central west coast of Taiwan. Results show that Mai-Liao Sand (MLS), a silty sand, can be significantly more compressible than clean quartz sand under static load. The particles of MLS have moderate strength, and significant crushing can be induced by triaxial shearing. As a result, MLS has low dilatancy and a relatively small range of peak friction angles. Cone penetration tests in MLS were conducted in a calibration chamber. Analyses of the data indicate that interpreting cone tip resistance in MLS using methods developed based on clean quartz sand without considering the differences of compressibility can be unrealistic. This paper documents results of the experimental studies on MLS.
The hydrodynamic coefficients for an oscillating rectangular structure on a free surface with sidewall
(1997-02-01)
Hsu, HH
;
Wu, YC
;
土木工程學系
;
Department of Civil Engineering
Based on a two dimensional linear water wave theory, the boundary element method (BEM) is developed and applied to study the heave and the sway problem of a floating rectangular structure in water to finite depth with one side of the boundary is a vertical sidewall and the other boundary is an open boundary. Numerical results for the added mass and radiation damping coefficients are presented. These coefficients are not only depend on the submergence and the width of the structure, but also depend on the clearance between structure and sidewall. Negative added mass and sharp peaks in the damping and added mass coefficients have been found when the clearance with a value close to integral times of half wave length of wave generated by oscillation structure. The important effect of the clearance on the added mass and radiation damping coefficients are discussed in detail. An analytical solution method is also presented. The BEM solution is compared with the analytical solution, and the comparison shows good agreement. Copyright (C) 1996 Elsevier Science Ltd