• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.11-18
• /
• 2012

Debris flow deposition model is a model to predict affected areas by debris flow and random walk model (RWM) was used to build the model. Although the model was proved to be effective in the prediction of affected areas, the model has several free parameters decided experimentally. There are several well-known methods to estimate parameters, however, they cannot be applied directly to the debris flow problem due to the small size of training data. In this paper, a modified neural network, called pseudo sample neural network (PSNN), was proposed to overcome the sample size problem. In the training phase, PSNN uses pseudo samples, which are generated using the existing samples. The pseudo samples smooth the solution space and reduce the probability of falling into a local optimum. As a result, PSNN can estimate parameter more robustly than traditional neural networks do. All of these can be proved through the experiments using artificial and real data sets.

• International Journal of Contents
• /
• v.8 no.3
• /
• pp.83-93
• /
• 2012

In this paper, we propose a novel method that automatically generates real character images to familiarize existing OCR systems with new fonts. At first, we generate synthetic character images using a simple degradation model. The synthetic data is used to train an OCR engine, and the trained OCR is used to recognize and label real character images that are segmented from ideal document images. Since the OCR engine is unable to recognize accurately all real character images, a substring matching method is employed to fix wrongly labeled characters by comparing two strings; one is the string grouped by recognized characters in an ideal document image, and the other is the ordered string of characters which we are considering to train and recognize. Based on our method, we build a system that automatically generates 2350 most common Korean and 117 alphanumeric characters from new fonts. The ideal document images used in the system are postal envelope images with characters printed in ascending order of their codes. The proposed system achieved a labeling accuracy of 99%. Therefore, we believe that our system is effective in facilitating the generation of numerous character samples to enhance the recognition rate of existing OCR systems for fonts that have never been trained.

• Korean Journal of Biological Psychiatry
• /
• v.18 no.3
• /
• pp.148-158
• /
• 2011

Objectives The eye movement (EM) has been reported to play a role in enhancing the retrieval of episodic memories and reducing effects of fearful episodes in the past and worries for the futures. However, it is still unclear in the mechanism of EM in normal subjects. We examined the horizontal eye movement (HEM) effect using an aiding apparatus on mental health indices including negative and positive psychological factors, and psychophysiological measures such as heart rate variability and quantitative electroencepaholography (qEEG) in healthy subjects. Methods Twenty eight healthy subjects were recruited and randomly allocated into two groups : active HEM group and control group. The active HEM group conducted the HEM training with usual stress management audio-intervention using the apparatus inducing eye movement once a day for 14 days. The control group also conducted the same training once a day for 14 days, however, the saccadic eye movement was not included in this training. Psychological measurements, neurocognitive function tests, heart rate variability measurement and qEEG were conducted before and after the training in both groups. Results In the active HEM group, sleep status using Sleep Quality Scale (SQS) positive factors significantly increased after the training. By contrast, scores on the negative items of Psychological Well-Being Scale (PWBS), and negative items of the Life Orientation Test-Revised (LOT-R) were significantly decreased after the training. The percentage of delta amplitude (1-3 Hz) in qEEG significantly decreased after the HEM training. The percentage of alpha amplitude (8-12 Hz) significantly increased after HEM training. The change of delta amplitude in the active HEM group was positively correlated with the change of sleep satisfaction of Visual Analogue Scale (VAS), and the change of alpha amplitude was negatively correlated with depression of VAS, anxiety of VAS and Beck Anxiety Inventory (BAI). Conclusions The HEM training improved sleep quality and well-being, and sense of optimism. The HEM training also increased alpha amplitude and decreased delta amplitude in qEEG. The qEEG changes were well correlated with subjective improvement of mental health indices in healthy subjects. These results suggest some evidences that HEM training using the apparatus that induces EM would be helpful in improving subjective mental health in healthy subjects. Further study with larger samples size would be needed.

• The Journal of the Korea Contents Association
• /
• v.13 no.11
• /
• pp.902-912
• /
• 2013

This study on the training of police officers job satisfaction and job performance is to identify the relationship. This study of 2012 Police Training Institute courses in related expenses 5 population in the process of initiation of a national police officer selection, and note that the sampling method used to extract a total of 300 samples, but the number of cases that were used in the final analysis, 268 people. Data processing by the SPSSWIN 18.0 factor analysis, reliability analysis, multiple regression, path analysis. Conclusions are as follows. First, the police officer's training affects job satisfaction. In other words, work-related, of course the more positive the evaluation of job training job satisfaction is high, education, the stronger the motivation and job satisfaction also higher education can be. Second, the education and training of police officers affects job performance. In other words, work-related, educational motivation, job training curriculum for the more positive job performance rating is also high. Third, the police officer's job satisfaction affects job performance. In other words, education can be a higher job performance and job satisfaction also high. Fourth, the training of police officers on the job satisfaction and job performance directly or indirectly affected. That is, the internal job satisfaction and job performance, job training parameters are the important variables.

• Asian-Australasian Journal of Animal Sciences
• /
• v.27 no.11
• /
• pp.1540-1547
• /
• 2014

The accuracy of genomic estimated breeding values (GEBV) was evaluated for sixteen meat quality traits in a Berkshire population (n = 1,191) that was collected from Dasan breeding farm, Namwon, Korea. The animals were genotyped with the Illumina porcine 62 K single nucleotide polymorphism (SNP) bead chips, in which a set of 36,605 SNPs were available after quality control tests. Two methods were applied to evaluate GEBV accuracies, i.e. genome based linear unbiased prediction method (GBLUP) and Bayes B, using ASREML 3.0 and Gensel 4.0 software, respectively. The traits composed different sets of training (both genotypes and phenotypes) and testing (genotypes only) data. Under the GBLUP model, the GEBV accuracies for the training data ranged from $0.42{\pm}0.08$ for collagen to $0.75{\pm}0.02$ for water holding capacity with an average of $0.65{\pm}0.04$ across all the traits. Under the Bayes B model, the GEBV accuracy ranged from $0.10{\pm}0.14$ for National Pork Producers Council (NPCC) marbling score to $0.76{\pm}0.04$ for drip loss, with an average of $0.49{\pm}0.10$. For the testing samples, the GEBV accuracy had an average of $0.46{\pm}0.10$ under the GBLUP model, ranging from $0.20{\pm}0.18$ for protein to $0.65{\pm}0.06$ for drip loss. Under the Bayes B model, the GEBV accuracy ranged from $0.04{\pm}0.09$ for NPCC marbling score to $0.72{\pm}0.05$ for drip loss with an average of $0.38{\pm}0.13$. The GEBV accuracy increased with the size of the training data and heritability. In general, the GEBV accuracies under the Bayes B model were lower than under the GBLUP model, especially when the training sample size was small. Our results suggest that a much greater training sample size is needed to get better GEBV accuracies for the testing samples.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.1
• /
• pp.199-206
• /
• 2019

This study aims to identify whether the action-observation training impacts on the improvement of stroke patient's cognitive functioning. When it comes to the research methods, Korean version of Mini-Mental State Examination (MMSE-K) and Literacy Independent Cognitive Assessment (LICA) conducted to assess samples between 23 April and 18 May 2018. Samples are seven patients who are hospitalized in Kyung-In region. In the meantime, seven tasks such as the range of joint motion (ROM) dance, arrangement of pullover clothes, lacing-ups of a pair, folding up a facecloth and socks, the origami and tying a necktie implemented as the action-observation programme. In order to analyse collected data, descriptive statistics analysis, paired t-test and the Wilcoxon signed-rank test were carried out via SPSS version 20 (a statistics programme). The change in value from MMSE-K showed its statistical significant as 3.29 (±1.38, p<.001) as well as value from LICA in recollective powers was 12.16 (±6.73), therefore, the statistic is said to be statistically significant. In conclusion, action-observation training most influenced recollective powers amongst stroke patient's cognitive functioning areas. Even though development of cognitive functioning discovered in other areas, its values were possibly statistically insignificant. Hence, future research ought to demonstrate which areas action-observation training is effective according to brain lesion site.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.1033-1038
• /
• 1994

This paper describes recognition results using the modified Learning Vector Quantization (MLVQ2) method which we proposed previously. At first, we investigated the property of duration of 29 Korean consonants and found that the variances of th duration were extremely big comparing to other languages. We carried out preliminary recognition experiments for three stop consonants P, T and K. From the recognition results, we defined the optimum conditions for the learning. Then we applied the MLVQ2 method to the recognition of Korean consonants. The training was carried out using the phoneme samples in the 611 word vocabulary uttered by 2 male speakers, where each of the speakers uttered two repetitions. The recognition experiment was carried out for the phoneme samples in two repetitions of the 611 word vocabulary uttered by another male speaker. The recognition scores for the twelve plosives were 68.2% for the test samples. The recofnition scores for the 29 Korean consonants were 64.8% for the test samples.

• Journal of Electrical Engineering and Technology
• /
• v.11 no.4
• /
• pp.848-859
• /
• 2016

System failures in thermal power plants (TPPs) can lead to serious losses because the equipment is operated under very high pressure and temperature. Therefore, it is indispensable for alarm systems to inform field workers in advance of any abnormal operating conditions in the equipment. In this paper, we propose a clustering-based fault detection method for steam boiler tubes in TPPs. For data clustering, k-means algorithm is employed and the number of clusters are systematically determined by slope statistic. In the clustering-based method, it is assumed that normal data samples are close to the centers of clusters and those of abnormal are far from the centers. After partitioning training samples collected from normal target systems, fault scores (FSs) are assigned to unseen samples according to the distances between the samples and their closest cluster centroids. Alarm signals are generated if the FSs exceed predefined threshold values. The validity of exponentially weighted moving average to reduce false alarms is also investigated. To verify the performance, the proposed method is applied to failure cases due to boiler tube leakage. The experiment results show that the proposed method can detect the abnormal conditions of the target system successfully.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.3962-3980
• /
• 2019

To deal with single sample face recognition, this paper presents a patch based semi-supervised linear regression (PSLR) algorithm, which draws facial variation information from unlabeled samples. Each facial image is divided into overlapped patches, and a regression model with mapping matrix will be constructed on each patch. Then, we adjust these matrices by mapping unlabeled patches to $[1,1,{\cdots},1]^T$. The solutions of all the mapping matrices are integrated into an overall objective function, which uses ${\ell}_{2,1}$-norm minimization constraints to improve discrimination ability of mapping matrices and reduce the impact of noise. After mapping matrices are computed, we adopt majority-voting strategy to classify the probe samples. To further learn the discrimination information between probe samples and obtain more robust mapping matrices, we also propose a multistage PSLR (MPSLR) algorithm, which iteratively updates the training dataset by adding those reliably labeled probe samples into it. The effectiveness of our approaches is evaluated using three public facial databases. Experimental results prove that our approaches are robust to illumination, expression and occlusion.

• The Korean Journal of Emergency Medical Services
• /
• v.12 no.1
• /
• pp.27-41
• /
• 2008

Purpose : With increasing number of school accidents, it is crucial to find out necessity of first aid training among school health educator. This study has been conducted to have an clear idea on the demands and necessity for first aid training and what kind of training is most required from school health educator. Method : In this study, questionnaires from 87 school health educator in elementary, middle and high school health educator in the city D were analyzed. The survey was carned out from May 26, 2008 to June 7, 2008 and from the collected data, frequency, independent two samples t-test, paired T-test, one way ANOVA and pearson's correlation were conducted with SPSS 14.0. Result: 1. 51.61 % of nurse-teachers experienced emergency situations and the relations between the necessity they felt from experiencing those situations and demands for first aid training were not statistically meaningful(t=1.87, p= .175). 2. Necessity and demands for the first-aid training were checked with three point scale and there were statistical significance between the two with $2.44{\pm}.47$ and $2.24{\pm}.47$ respectively(t=3.275, p= .000). 3. 86.20%(75 persons) of the respondents have had received first aid training and the training they received were CPR 82.75%(72 persons), primary survey 81.60%(7l persons), contact to 911 79.30%(69 persons) and wounds treatment(lacerated wounds, bum and chilblains) 75.86%( 66 persons) in order. 4. As for the questions that ask on confidence of first-aid treatment, 80% answered they are confident on some limited kinds of treatments, 16% said they are confident and 4% answered they lack confidence. As for the treatment that they can show the highest confidence, wounds treatment topped the list with 93.24%, nose bleeding and removing foreign substance, and stanching followed the list with 82.43% and 81.08% respectively. 5. 97.67% of respondents said they were willing to take training and 89.62% answered to take the training to deal with emergency situations that are taking place in their schools. As for the question that asks for the most wanted treatments, CPR topped the list with 32.18%(28 persons) and treatment for obstruction of airway and shock followed the list with 35.63%(31 persons) and 27.59%(24 persons). Conclusion : Currently, first aid treatment has been centered on CPR, primary survey, contact to 119 and wounds treatment. However, since most of school health educator are fairly confident with wounds treatment, stanching and other first aids, in future training it will be more desirable to focus on CPR and treatment for obstruction of airway and shock that were shown to be most wanted by school health educator.