• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.4
• /
• pp.23-30
• /
• 2012

This paper proposes an improved steganalysis method to detect the existence of secret message. Firstly, we magnify the small stego noise by multiplying the speckle noise to a given image and then we estimate the denoised image by using the soft thresholding method. Because the noises are not perfectly eliminated, some noises exist in the estimated cover image. If the given image is the cover image, then the remained noise will be very small, but if it is the stego image, the remained noise will be relatively large. The parent-child relationship in the wavelet domain will be slighty broken in the stego image. From this characteristic, we extract the joint statistical moments from the difference image between the given image and the denoised image. Additionally, four statistical moments are extracted from the denoised image for the proposed steganalysis method. All extracted features are used as the input of MLP(multilayer perceptron) classifier. Experimental results show that the proposed scheme outperforms previous methods in terms of detection rates and accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.24-31
• /
• 1994

In this paper, we report on the development of a speaker independent continuous speech recognition system using continuous hidden Markov models. The continuous hidden Markov model consists of mean and covariance matrices and directly models speech signal parameters, therefore does not have quantization error. Filter bank coefficients with their 1st and 2nd-order derivatives are used as feature vectors to represent the dynamic features of speech signal. We use the segmental K-means algorithm as a training algorithm and triphone as a recognition unit to alleviate performance degradation due to coarticulation problems critical in continuous speech recognition. Also, we use the one-pass search algorithm that Is advantageous in speeding-up the recognition time. Experimental results show that the system attains the recognition accuracy of $83\%$ without grammar and $94\%$ with finite state networks in speaker-indepdent speech recognition.

• The KIPS Transactions:PartD
• /
• v.12D no.4 s.100
• /
• pp.647-656
• /
• 2005

This paper proposes an adaptive resource self-management system that collects system resources, user information, and usage patterns as context information for utilization in self-configuration. This system ill ease the system maintenance burden on users by automation of large part of configuration tasks such as install, reconfiguration and update, and will also decrease cost and errors. Working from the gathered context information, this system allows users to select appropriate components and install them for user's system context. This also offers a more personalized configuration setting by using user's existing application setting and usage pattern. To avoid the overload on central server to transfer and manage related files, we employ Peer-to-Peer method. h prototype was developed to evaluate the system and a comparison study with the conventional methods of manual configuration and MS-IBM systems was conducted to validate the proposed system in terms of functional capacity, install time and etc.

• Journal of the Korean Chemical Society
• /
• v.45 no.5
• /
• pp.422-428
• /
• 2001

Retention relative times(RRTs) of PAH molecules and their derivatives in gas chromatography are trained and predicted in testing sets using a multiple linear regression(MLR) and an artificial neural network(ANN). The main descriptors of PAHs and their derivatives in QSRR are the square root of molecular weight(sqmw), molecular connectivity($^1{\chi}_v$), molecular dipole moment(D) and length-to-breadth ratios(L/B). The results of MLR shows that a heavy molecule has a propensity for long retention time. L/B closely related with slot model is a good descriptor in MLR. On the other hand, ANN which is not effected by the linear dependencies among the descriptors were exclusively based on molecular weight and molecular dipole moment. The variances which shows the accuracy of prediction for retention times in testing sets are 1.860, 0.206 for MLR and ANN, respectively. It was shown that ANN can exceed the MLR in prediction accuracy.

• The Journal of Korean Physical Therapy
• /
• v.26 no.4
• /
• pp.234-239
• /
• 2014

Purpose: The purpose of this study is to use P300 latency to determine whether methods of motor learning in terms of massed and distributed practice can affect motor sequential learning in healthy adults. Methods: Twenty-four healthy subjects participated in this study. They were randomly allocated into three groups: a 10 minute, a 12 hour, and a 24 hour group. In the SRT task, eight numbers were adopted as auditory stimuli. During an experiment, participants were instructed to press the matching key as quickly and accurately as possible when one of the eight numbers was presented randomly. The subjects practiced for three sessions, each of which comprised five blocks of 40 serial reaction time tasks. While they practiced during these three sessions, P300 latency was measured. The data were analyzed using ANCOVA. Results: The P300 latency of Fz, Cz, and Pz decreased in all groups except for the Fz area of the 10 min group. Overall, the P300 latency of the 10 min group showed a smaller decrease compared with the 12 hr and 24 hr groups. Statistically, no significant differences in the Fz and Cz areas were observed among the three groups. The P300 latency in the Pz area of the 10 min group showed a significantly smaller decrease compared with the other groups. Conclusion: These findings suggest that short-term sequential motor training can alter brain functions such as the P300 latency. We also found that better acquisition of a motor skill was obtained with distributed practice of a task than with massed practice.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.10
• /
• pp.166-174
• /
• 1998

Neural networks, having highly nonlinear dynamics by virtue of the distributed nonlinearities and the learing ability, have the potential for the adaptive prediction of nonstationary signals. This paper describes the nonlinear prediction of these signals in two ways; using a nonlinear module and the cascade combination of nonlinear and linear modules. Fully-connected recurrent neural networks (RNNs) and a conventional tapped-delay-line (TDL) filter are used as the nonlinear and linear modules respectively. The dynamic behavior of the proposed predictors is demonstrated for chaotic time series adn speech signals. For the relative comparison of prediction performance, the proposed predictors are compared with a conventional ARMA linear prediction model. Experimental results show that the neural networks based adaptive predictor ourperforms the traditional linear scheme significantly. We also find that the cascade combination predictor is well suitable for the prediction of the time series which contain large variations of signal amplitude.

• Korean Journal of Social Welfare
• /
• v.65 no.1
• /
• pp.5-31
• /
• 2013

The principle objective of this study is to determine factors affecting social participation for physically disabled people with musculoskeletal conditions (spinal cord injury, muscular dystrophy, osteogenesis imperfecta, rheumatoid arthritis) in South Korea using ICF relevant categories of the international classification index by WHO. The subjects of this study 352 people with physical disabilities, the data were collected using ICF component(body functions, body structures, activities and participation and environmental factors) and the relevant categories deprived from ICF core sets. The collected data were evaluated with descriptive analysis, ANOVA, correlation analysis, and multiple regression analysis. The results of this study can be summarized as follow. The mental function, Neuromusculoskeletal and movement-related functions, Genitourinary and reproductive functions, Skin and related structures, Learning and applying knowledge, General tasks and demands, and Mobility positively influenced social participation in people with musculoskeletal conditions. However individual factors and environmental factors didn't statistically significant affect on social participation. The implications of the study is to examine by ICF components of universal approach on disability study and utilized the relevant ICF categories as measurement tools.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.7
• /
• pp.540-547
• /
• 2004

Since speech enhancement algorithms based on Kalman/Wiener filter require a priori knowledge of the noise and have focused on the minimization of the variance of the estimation error between clean and estimated speech signal, small estimation error on the noise statistics may lead to large estimation error. However, H/sub ∞/ filter does not require any assumptions and a priori knowledge of the noise statistics, but searches the best estimated signal among the entire estimated signal by applying least upper bound, consequently it is more robust to the variation of noise statistics than Kalman/Wiener filter. In this paper, we Propose a speech enhancement method using HMM and multi H/sub ∞/ filters. First, HMM parameters are estimated with the training data. Secondly, speech is filtered with multiple number of H/sub ∞/ filters. Finally, the estimation of clean speech is obtained from the sum of the weighted filtered outputs. Experimental results shows about 1dB∼2dB SNR improvement with a slight increment of computation compared with the Kalman filter method.

• Education of Primary School Mathematics
• /
• v.10 no.2
• /
• pp.125-139
• /
• 2007

The purpose of this study seeks entry into a method to make the use of educational aids popular. To achieve it, it is observed that instructions applying worksheets to make an activation of use of educational aids have influences on mathematical achievement and mathematical disposition and attitude. All variables exception with the frequence of use of educational aids are controlled in both experimental group and comparative group. According to the result, there is no significant difference of mathematical achievement in pre t-test between two groups, while experimental group get 10 points higher than comparative group in average (t=0.519, p<0.01). On the other hand, within intra-experimental group the influences of use of educational aids on mathematical achievement is positive without the achievement levels of students. The difference dependent on the levels of student is sought by ANCOVA using prescores as a covariance, and it appears in the significance level of 5%(F=4.885, p<0.05), and the effect is more in the lower level of students than in the middle and high level.

• The KIPS Transactions:PartD
• /
• v.8D no.6
• /
• pp.673-680
• /
• 2001

When the developers develop the software, the cost and time of the software development can be reduced by using blocks that are implemented previously. We call these implemented blocks components. In the early stage of Web-based Instruction, it didn't gain preference in spite of it's benefit of convenience. The main reason is, I think, the lack of generality at the education system which eventually results in unsatisfactory facilities compared with the requirement of teachers and students. And the early systems don't make good use of the plenty data in distributed environment, and don't show so good reliablity due to lack of systematic design and development. In this paper, we suggest WBI developing technology using the concept of WBSE. WBI developing is consist of component of pre-developed education software, integration of component using its reusability, and production of more requirement-satisfactory education software.