• Phonetics and Speech Sciences
• /
• v.2 no.2
• /
• pp.63-68
• /
• 2010

A new version of sequential probability ratio test (SPRT) which has been investigated in utterance-length control is proposed to obtain uniform response results in speaker verification (SV). Although SPRTs can obtain fast responses in SV tests, differences in the performance may occur depending on the compositions of consonants and vowels in the sentences used. In this paper, a fast sequential probability ratio test (FSPRT) method that shows consistent performances at all times regardless of the compositions of vocalized sentences for SV will be proposed. In generating frames, the FSPRT will first conduct SV test processes with only generated frames without any overlapping and if the results do not satisfy discrimination criteria, the FSPRT will sequentially use frames applied with overlapping. With the progress of processes as such, the test will not be affected by the compositions of sentences for SV and thus fast response outcomes and even consistent performances can be obtained. Experimental results show that the FSPRT has better performance to the SPRT method while requiring less complexity with equal error rates (EER).

• Nuclear Engineering and Technology
• /
• v.56 no.2
• /
• pp.715-727
• /
• 2024

During fast neutron imaging, besides the dark current noise and readout noise of the CCD camera, the main noise in fast neutron imaging comes from high-energy gamma rays generated by neutron nuclear reactions in and around the experimental setup. These high-energy gamma rays result in the presence of high-density gamma white spots (GWS) in the fast neutron image. Due to the microscopic quantum characteristics of the neutron beam itself and environmental scattering effects, fast neutron images typically exhibit a mixture of Gaussian noise. Existing denoising methods in neutron images are difficult to handle when dealing with a mixture of GWS and Gaussian noise. Herein we put forward a deep learning approach based on the Swin Transformer UNet (SUNet) model to remove high-density GWS-Gaussian mixture noise from fast neutron images. The improved denoising model utilizes a customized loss function for training, which combines perceptual loss and mean squared error loss to avoid grid-like artifacts caused by using a single perceptual loss. To address the high cost of acquiring real fast neutron images, this study introduces Monte Carlo method to simulate noise data with GWS characteristics by computing the interaction between gamma rays and sensors based on the principle of GWS generation. Ultimately, the experimental scenarios involving simulated neutron noise images and real fast neutron images demonstrate that the proposed method not only improves the quality and signal-to-noise ratio of fast neutron images but also preserves the details of the original images during denoising.

• Advances in materials Research
• /
• v.9 no.2
• /
• pp.115-131
• /
• 2020

Limit elastic speed analysis of Al/SiC-based functionally graded annular disk of uniform thickness has been carried out for two cases, namely: metal-rich and ceramic rich. In the present study, the unknown field variable for radial displacement is solved using variational method wherein the solution was obtained by Galerkin's error minimization principle. One of the objectives was to identify the variation of induced stress in a functionally graded disk of uniform thickness at limit elastic speed using modified rule of mixture by comparing the induced von-Mises stress with the yield stress along the disk radius, thereby locating the yield initiation. Furthermore, limit elastic speed has been reported for a combination of varying grading index (n) and aspect ratios (a/b).Results indicate, limit elastic speed increases with an increase in grading indices. In case of an increase in aspect ratio, limit elastic speed increases up to a critical value beyond which it recedes. Also, the objective was to look at the variation of yield stress corresponding to volume fraction variation within the disk which later helps in material tailoring. The study reveals the qualitative variation of yield stress for FG disk with volume fraction, resulting in the possibility of material tailoring from the processing standpoint, in practice.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.688-694
• /
• 2018

The mixture ratio of two-component liquid silicone is important for the inherent physical characteristics of the finished product. Therefore, it is necessary to uniformly control the ratio of the main material and the sub-material. In this paper, a mixing-ratio control system was designed, which consists of a digital flow meter and a flow control system to measure the flow rate of the raw materials and a pumping system to maintain constant pressure and transfer of the raw materials. In addition, a program was developed to control the organic interlocking and mixing ratio. For the verification of the developed system, we compared the actual weight of raw material with the value measured by the flow meter during pumping, and we measured the physical properties of the mixed material by making test samples with and without the application of the mixing-ratio improvement algorithm. The measured value was close to the reference value with a hardness range of 46-47 and tensile strength of 9.3-9.5 MPa. These results show that the mixing ratio of the liquid silicone is controlled within an error range of ${\pm}0.5%$.

• The Korean Journal of Applied Statistics
• /
• v.23 no.1
• /
• pp.13-28
• /
• 2010

Assuming a mixture distribution for credit evaluation studies, we discuss estimating threshold methods to minimize errors that default borrowers are predicted as non defaults or non defaults are regarded as defaults. A method by using statistical hypotheses tests, the most powerful test and generalized likelihood ratio test, for the probability density functions which are defined with the score random variable and the parameter space consisted of only two elements such as the default and non default states is proposed to estimate a threshold. And anther optimal thresholds to maximize classification accuracy measures of the accuracy and the true rate for ROC and CAP curves are estimated as equations related with these probability density functions. Three kinds of optimal thresholds in terms of the hypotheses testing, the accuracy and the true rate are obtained from normal random samples with various means and variances. The sums of the type I and type II errors corresponding to each optimal threshold are obtained and compared. Finally we discuss about their efficiency and derive conclusions.

• Advances in concrete construction
• /
• v.10 no.5
• /
• pp.381-391
• /
• 2020

This study aimed to develop reference materials (RMs) for mortar that can simulate the initial flow characteristics with constant quality over a long period. Through the previous research on the development of RMs for cement paste, the combination of limestone, glycerol, and water was used as the basic matrix for developing RMs for mortar in this study. In addition, glass beads of three particle sizes (0.5, 1.0, and 2.0 mm) and ISO standard sand were selected as tentative candidates to derive fine aggregate substitutes. The mixture of glass beads could simulate the initial flow characteristics of mortar, but under the same mixing ratio, replicates showed an unstable tendency to indicate inconsistent values due to the generation of electrostatic properties between materials and equipment. On the other hand, the mixture using ISO standard Sand not only simulates the constant flow characteristics for a long period of time, but also shows stable results with little error in replicates. Therefore, limestone, glycerol, ISO standard sand, and water were finally determined as components that met the required properties of RMs for mortar. The effect of each component on the flow characteristics of RMs was analyzed. It was found that glycerol increased the cohesion between the particles of standard sand, resulting in a constant increase both in the plastic viscosity and yield stress. Both limestone and standard sand had a dominant effect on the yield stress. The relationships between various mortar mixing ratios and the corresponding mixing ratios of RMs were established. In addition, the results of the verification experiment showed that the rheological properties of the RMs obtained through the relationships correlated with various water/cement ratios and the fine aggregate volume fractions of mortar obtained with same manner. In other words, the RMs for mortar developed in this study can be used as standard samples because they can simulate the initial flow characteristics of mortar of various mixing ratios for a long period without any chemical changes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.14 no.3
• /
• pp.280-292
• /
• 1989

In this paper, the error rate equations of digitally modulated signals transmitted through the channel which is not only Gaussian/Impulsive noise but also multi-path fading have been derived. Using the derived equations for the error probabilities of ASK, QAM, CPSK, DPSK, FSK, and MSK signals, the error rate performances of digital modulation systems have been evaluated and represented in the graphs as the functions of CNR, Impulsive indes, the ratio of Gaussian noise power component to Impulsive noise power component, and fading figures. The results show that, in the deep fading environment, the error is occurred more frequency by Gaussian noise than Impulsive noise. And the comparison of various systems certifies that PSK is superior to the ohter systems in the deep fading or shallow fading environment.

• MALSORI
• /
• no.67
• /
• pp.195-216
• /
• 2008

This paper addresses the problem of single-channel speech separation to extract the speech signal uttered by the speaker of interest from a mixture of speech signals. We propose to apply time-frequency smoothing to the existing statistical single-channel speech separation algorithms: The soft mask and the minimum-mean-square-error (MMSE) algorithms. In the proposed method, we use the two smoothing later. One is the uniform mask filter whose filter length is uniform at the time-Sequency domain, and the other is the met-scale filter whose filter length is met-scaled at the time domain. In our speech separation experiments, the uniform mask filter improves speaker-to-interference ratio (SIR) by 2.1dB and 1dB for the soft mask algorithm and the MMSE algorithm, respectively, whereas the mel-scale filter achieves 1.1dB and 0.8dB for the same algorithms.

• International Journal of Aerospace System Engineering
• /
• v.6 no.2
• /
• pp.11-16
• /
• 2019

A system analysis program has been developed for a gas generator cycle liquid rocket engine of 30 ton class. Numerical models have been proposed for a combustor, a turbopump, a gas generator and pressure drop through a regenerative cooling system. Numerical algorithm has been validated by comparing with the published data of MC-1. The major source of error is not the numerical algorithm but the imperfect performance models of subsystems. So the precision of the program can be improved by revising the performance models using experimental data. The sea level specific impulse and vacuum specific impulse have been demonstrated for a 30 ton class gas generator engine. The optimal condition of combustor pressure and mixture ratio for specific impulse which is a typical characteristic of a gas generator cycle engine has been illustrated.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.402-404
• /
• 2003

We present two methods for the automatic selection of the threshold values in unsupervised change detection. Both methods consist of the same two procedures: 1) to determine the parameters of Gaussian mixtures from a difference image or ratio image, 2) to determine threshold values using the Bayesian rule for minimum error. In the first method, the Expectation-Maximization algorithm is applied for estimating the parameters of the Gaussian mixtures. The second method is based on the iterative thresholding that successively employs thresholding and estimation of the model parameters. The effectiveness and applicability of the methods proposed here are illustrated by an experiment on the multi-temporal KOMPAT-1 EOC images.