• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.189-198
• /
• 2024

Lipreading is one of the important parts of speech recognition, and several studies have been conducted to improve the performance of lipreading in lipreading systems for speech recognition. Recent studies have used method to modify the model architecture of lipreading system to improve recognition performance. Unlike previous research that improve recognition performance by modifying model architecture, we aim to improve recognition performance without any change in model architecture. In order to improve the recognition performance without modifying the model architecture, we refer to the cues used in human lipreading and set other regions such as chin and cheeks as regions of interest along with the lip region, which is the existing region of interest of lipreading systems, and compare the recognition rate of each region of interest to propose the highest performing region of interest In addition, assuming that the difference in normalization results caused by the difference in interpolation method during the process of normalizing the size of the region of interest affects the recognition performance, we interpolate the same region of interest using nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation, and compare the recognition rate of each interpolation method to propose the best performing interpolation method. Each region of interest was detected by training an object detection neural network, and dynamic time warping templates were generated by normalizing each region of interest, extracting and combining features, and mapping the dimensionality reduction of the combined features into a low-dimensional space. The recognition rate was evaluated by comparing the distance between the generated dynamic time warping templates and the data mapped to the low-dimensional space. In the comparison of regions of interest, the result of the region of interest containing only the lip region showed an average recognition rate of 97.36%, which is 3.44% higher than the average recognition rate of 93.92% in the previous study, and in the comparison of interpolation methods, the bilinear interpolation method performed 97.36%, which is 14.65% higher than the nearest neighbor interpolation method and 5.55% higher than the bicubic interpolation method. The code used in this study can be found a https://github.com/haraisi2/Lipreading-Systems.

• Korean Journal of Social Welfare
• /
• v.60 no.2
• /
• pp.77-98
• /
• 2008

For women living with mental illness, motherhood may be a normalizing life experiences and offers the opportunity to develop competencies in a major life roles. So, the main premise of this study that social contexts, social support of spouse and extended family, is very important for successful parenting of mothers with mental illness. Thus this study analyzes the impact of marital relationship and social support of networks on three dimensions of parenting competency(expression of love, encouragement of independence, rational discipline) when the socioeconomic(age, number of offsprings, subjective economic level) and disability variables(diagnosis, severity of symptoms, chronicity, frequencies of psychiatric admission) are controlled. This study conducted the survey with 145 mothers with schizophrenia and mood disorder. The major finding of the study are : i) Using step-wise multiple regression, marital relationship, social supports and economic level affect significantly on the expression of love. ii) Using step-wise multiple regression, the most powerful influenced factor on encouragement of independence is marital relationship. iii) Using step-wise multiple regression, diagnosis, severity of symptoms, chronicity and economic level affect significantly on the rational discipline. That is two dimensions of parenting competency, expression of love and encouragement of independence, can be explained well by social contexts variables(marital relationship, social support, economic level). But only rational discipline can be explain well by disability factors. In conclusion, this study recommends service providers to use family based comprehensive case management for the mothers with mental illness.

• Asia pacific journal of information systems
• /
• v.18 no.3
• /
• pp.1-20
• /
• 2008

Based on the multi-attribute utility theory (MAUT), we present a survey method to measure consumer preferences. The multi-attribute utility theory has been used to make decisions in OR/MS field; however, we show that the method can be effectively used to estimate the demand for new services by measuring individual level utility function. Because conjoint method has been widely used to measure consumer preferences for new products and services, we compare the pros and cons of two consumer preference survey methods. Further, we illustrate how swing weighing method can be effectively used to elicit customer preferences especially for new telecommunications services, Multi-attribute utility theory is a compositional approach for modeling customer preference, in which researchers calculate overall service utility by summing up the evaluation results for each attribute. On the contrary, conjoint method is a decompositional approach, which requires holistic evaluations for profiles. Partworth for each attribute is derived or estimated based on the evaluation, and finally consumer preferences for each profile are calculated. However, if the profiles are quite new and unfamiliar to the survey respondents, they will find it very difficult to accurately evaluate the profiles. We believe that the multi-attribute utility theory-based survey method is more appropriate than the conjoint method, because respondents only need to assess attribute level preferences and not holistic assessment. We chose swing weighting method among many weight assessment methods in multi-attribute utility theory, because it is designed to perform in a simple and fast manner. As illustrated in Clemen and Reilly (2001), to assess swing weights, the first step is to create the worst possible outcome as a benchmark by setting the worst level on each of the attributes. Then, each of the succeeding rows "swings" one of the attributes from worst to best. Upon constructing the swing table, respondents rank order the outcomes (rows). The next step is to rate the outcomes in which the rating for the benchmark is set to be 0 and the rating for the best outcome to be 100, and the ratings for other outcomes are determined in the ranges between 0 and 100. In calculating weight for each attribute, ratings are normalized by the total sum of all ratings. To demonstrate the applicability of the approach, we elicited and analyzed individual-level customer preference for new telecommunication services-WiBro and HSDPA. We began with a randomly selected 800 interviewees, and reduced them to 432 because other remaining ones were related to the people who did not show strong intention for subscription to new telecommunications services. For each combination of content and handset, number of responses which favored WiBro and HSDPA were counted, respectively. It was assumed that interviewee favors a specific service when expected utility is greater than that of competing service(s). Then, the market share of each service was calculated by normalizing the total number of responses which preferred each service. Holistic evaluation of new and unfamiliar service is a tough challenge for survey respondents. We have developed a simple and easy method to assess individual level preference by estimating weight of each attribute. Swing method was applied for this purpose. We believe that estimating individual level preference will be quite flexibly used to predict market performance of new services in many different business environments.

• Journal of Ginseng Research
• /
• v.15 no.3
• /
• pp.171-178
• /
• 1991

The results ok feeding experiments to the mice with ginseng extract, ginseng Powder, and ADAPTAGEN, for 30 days before X-ray irradiation and for 40 days after the X-ray irradiation at 750 rads were as follow: 1. The 50% lethals days (LD50, ) by the X-ray irradiation were 9 days at 1, 000 rads. 10 days at 900 rads, 11 days at 800 rads, 14 days at 760 rads, and 19 darts at 750 rads. Therefore, the standard radiation dose was set at 750 radb/8 min. 2. The 80% of the control group mice exposed to the X-ray radiation without ginseng feeding died in periods ranging from 14 to 24 days and the 20~30% of the ginseng extract and ginseng powder feeding groups died. But the 100% of the mice fed with ADAPTAGEN survived. 3. Testicles of the control group became smaller in weight than the nomad group by 26.5 to 29.0% and those of the ginseng extract and ginseng powder feeding group reduced by 44.6 to 60.4%. However, testicles of the ADAPTiIGEN feeding group increased in size by 77.4% to 87.1% and in weight by 61%, showing a recovery phenomenon approarhing to those of the ordinary mice. The ADAPTAGEN feeding group mice were also as active in color as the ordinary ones. 4. An electron micrograph(X8, 000X2.2) of the liver cells of the mice which had been 40 days after X-ray irradiation showed as follows; The control group appeared that is physiological action stopped due to the frequent occurrence of morphological change of the nucleus and diffusion of chromosome, reduction in microspores and expansion of microsomts, and endoplasmic change of mitochondria. The liver cells or the ADAPTAGEN feeding group were in a state similar to those of the ordinary mice restoring to normalcy In contrast, the liver cells of the ginseng extract and ginseng powder feeding groups were still far from being normal. 5. A serological analysis showed that the control group sharply decreased in albumin, Y-g1obu1in, and IgG so far as to cause dystrophy and to weaken antibody resistance but that ginseng extract and ginseng powder feeding groups, though in a little more restoring state than the control group, were still far from the normal group. The ADAPTAGEN feeding group restored to a state as comparable to the normal group in the contents of albumin ${\gamma}$-globulin, IgG and serum protein. In order words, it is noteworthy that ADAPTAGEN feeding was effective in revitalizing the destroyed cells of a living body and that it has the function of normalizing antibody components.

• Journal of Embryo Transfer
• /
• v.24 no.3
• /
• pp.225-230
• /
• 2009

Sex-sorting of sperm is an assisted reproductive technology (ART) used by the livestock industry for the mass production of animals of a desired sex. The standard method for sorting sperm is the detection of DNA content differences between X and Y chromosome-bearing sperm by flow cytometry. However, this method has variable efficiency and therefore requires verification by a second method. We have developed a sex determination method based on quantitative real-time polymerase chain reaction (qPCR) of the porcine amelogenin (AMEL) gene. The AMEL gene is present on both the X and the Y chromosome, but the length and sequence of its noncoding regions differ between the X and Y chromosomes. By measuring the threshold cycle (Ct) of qPCR, we were able to calculate the relative frequency of X chromosome. Two sets of AMEL primers were used in these studies. One set (AME) targeted AMEL gene sequences present in both X and Y chromosome, but produced PCR products of different lengths for each chromosome. The other set (AXR) bound to AMEL gene sequences present on the X chromosome but absent esholthe Y-chromosome. Relative product levels were calculated by normalizing the AXR fluorescence to the AME fluorescence. The AMEL method accurately predicted the sex ratios of boar sperm, demonstrating that it has potential value as a sex determination method.

• Journal of Digital Contents Society
• /
• v.1 no.1
• /
• pp.53-64
• /
• 2000

This paper is related to color image segmentation scheme which makes it possible to achieve the excellent segmented results by block-based segmentation using Y/C bit-plane summation image. First, normalized chrominance summation image is obtained by normalizing the image which is summed up the absolutes of color-differential values between R, G, B images. Secondly, upper 2 bits of the luminance image and upper 6bits of and the normalized chrominance summation image are bitwise operated by the pixel to generate the Y/C bit-plane summation image. Next, the Y/C bit-plane summation image divided into predetermined block size, is classified into monotone blocks, texture blocks and edge blocks, and then each classified block is merged to the regions including one more blocks in the individual block type, and each region is selectively allocated to unique marker according to predetermined marker allocation rules. Finally, fine segmented results are obtained by applying the watershed algorithm to each pixel in the unmarked blocks. As shown in computer simulation, the main advantage of the proposed method is that it suppresses the over-segmentation in the texture regions and reduces computational load. Furthermore, it is able to apply global parameters to various images with different pixel distribution properties because they are nonsensitive for pixel distribution. Especially, the proposed method offers reasonable segmentation results in edge areas with lower contrast owing to the regional characteristics of the color components reflected in the Y/C bit-plane summation image.

• Journal of Korea Water Resources Association
• /
• v.37 no.7
• /
• pp.553-567
• /
• 2004

This study proposes a modified standardized precipitation index (MSPI) which was developed to make up for the weakness of the SPI. Both MSPI and SPI are applied to the monthly rainfall at the Seoul station for the drought analysis. The MSPI proposed is nothing but the SPI for the normalized monthly rainfall, that is, an extra step for normalizing the monthly rainfall is included before driving the SPI. Thus, the MSPI has a structure to transfer the relative amount of rainfall to the next months, but the SPI the absolute amount of rainfall. The monthly rainfall data at the Seoul station used in this study are those collected from 1777 to 1996. The rainfall data collected before and after the long dry period around 1900 were also analyzed separately for the comparison. The results derived are as follows. (1) The MSPI was found to be more practical compared to the SPI. This was assured by comparing the analysis results of the data including and excluding the long dry period around 1900. (2) The MSPI is found to be less sensitive than the SPI to the extreme rainfall events. For the MSPI, the occurrence probabilities of moderate drought before and after the long dry period are similar, but those for the extreme drought becomes slightly decreased after the long dry period (from about 18 years of return period before the long dry period to the 16 years after the long dry period). However, the duration becomes longer after the long dry period (the duration for the extreme drought has been increased from 2 to 2.5 months after the long dry period). This results can also be compared with a rather unreasonable result derived by applying the SPI (for the extreme drought the return period has been decreased to be from 25 to 10 years after the long dry period, on the other hand the duration has been increased from 1.5 months to 3.5 months). So, we man conclude that the MSPI is more practical for the drought analysis that the SPI.

• Korean Journal of Acupuncture
• /
• v.24 no.4
• /
• pp.99-110
• /
• 2007

Objectives : Epilepsy is one of the most common serious brain disorders that affect people of all ages, and it is characterized by recurrent unprovoked seizures. We examined whether acupuncture can reduce both the incidence of seizures and hippocampal cell death in dentate gyrus (DG) using a mouse model of kainic acid (KA)-induced epilepsy. Methods : ICR mice ($20{\sim}25$ g) were given acupuncture once a day at acupoint HT8 (sobu) bilaterally during 2 days before KA injection. After an intracerebroventricular injection of 0.1${\mu}g$ of KA, acupuncture treatment was subsequently administered once more (total 3 times), and the degree of seizure was observed for 20 min. Three hours after injection, we confirmed the neural cell death using cresyl violet staining and silver impregnation staining, and determined the expressions of c-Fos and glutamate decarboxylase (GAD)-67 using immunohistochemistry techniques in the DG. Results : KA induced epileptic seizure, neural cell death, increased c-Fos expression and decreased GAD-67 expression in the DG. Acupuncture treatment at HT8 reduced the severity of the epileptic seizure and inhibited neural cell death from KA. In addition, acupuncture normalized the expressions of c-Fos and GAD-67 in the same areas. Conclusions : These results demonstrated that acupuncture treatment at HT8 may reduce the KA-induced epileptic seizure and neural cell death in the DG possibly by normalizing c-Fos expressions and the gamma-aminobutyric acid neurons.

• Journal of Wetlands Research
• /
• v.13 no.3
• /
• pp.411-426
• /
• 2011

The construction of composite indicators should be normalized and weighted to render them comparable and evaluable variables in the field, which undergoes absence of a distinct methodology and where the application of universally popular method is common. Constructing of indices does not compare and analyze applying various normalizing and weighting, but constructer generally use chosen method and develops indicators and indices in most research. In this study, indices are applied various normalization and weighting methods, thereby analyzing how much impact the index and identifying individual characteristics derive a more reasonable way to help other research in the future. 5 different methods of normalization and 4 different types of weights were compared and analyzed. There are different results depending applied normalized methods and Z-score method best reflects the characteristics of the variables. According to weighting methods, the calculated results show little difference, but the ranking results of indices did not changed significantly. It might be better to provide constructors with a set of normalization and weighting methods to reflect their characteristics in order to build flood indices through the result of this study.

• Investigative Magnetic Resonance Imaging
• /
• v.15 no.2
• /
• pp.110-122
• /
• 2011

Purpose : The purpose of this study is to establish the method generating human brain anatomical connectivity from Korean children and evaluating the network topological properties using small-world network analysis. Materials and Methods : Using diffusion tensor images (DTI) and parcellation maps of structural MRIs acquired from twelve healthy Korean children, we generated a brain structural connectivity matrix for individual. We applied one sample t-test to the connectivity maps to derive a representative anatomical connectivity for the group. By spatially normalizing the white matter bundles of participants into a template standard space, we obtained the anatomical brain network model. Network properties including clustering coefficient, characteristic path length, and global/local efficiency were also calculated. Results : We found that the structural connectivity of Korean children group preserves the small-world properties. The anatomical connectivity map obtained in this study showed that children group had higher intra-hemispheric connectivity than inter-hemispheric connectivity. We also observed that the neural connectivity of the group is high between brain stem and motorsensory areas. Conclusion : We suggested a method to examine the anatomical brain network of Korean children group. The proposed method can be used to evaluate the efficiency of anatomical brain networks in people with disease.