• Korean Journal of Animal Reproduction
• /
• v.8 no.1
• /
• pp.46-55
• /
• 1984

The purpose of the experiment was to clarify morphologically normal growth pattern of the ductus deference in accordance with the sex maturity of meat-type cockerels. 1. Diameter of lumens in u, pp.r, mid and lower parts of ductus deferens, the most conspicuous enlargement of lumen was observed in the lower part. Heights of epithelial layers of ductus deferens showed abrupt growth at 12 weeks of age with subsequent gradual growth in all the part of u, pp.r, mid and lower, and heights of those at 30 weeks were a, pp.oximately 4 times as large in the u, pp.r and mid parts and 5 times as large in the lower part in contrast to those at 4 weeks of age. Thickness of muscular layer of ductus showed gradual growth in contrast with the diameter of lumen and height of epithelial layer, showing 1.3 times as large in the u, pp.r part, 1.6 times in the mid part and 1.9 times in the lower part at 30 weeks of age in contrast to the thickness at 4 weeks of age. 2. Within 10 weeks after hatching, lining cells of ductus deferens were mainly composed of round cells and columnar cells in simple columnar epithelium. During 10th to 20th week, the lining cells were mainly composed of high columnar cells and round cells in pseudostratified epithelium. From 22nd week, the lining cells were composed of pseudostratified columnar cells. Whereas round cells disa, pp.ared gradually. Enlargement of lumen and pooling of sperms in ductus deferens coincided with the maturation of seminiferous tubules. 3. In simple correlation between the values of testis weight and the values from various measurements in the ductus deferens, there was significant correlation coefficient with each other. 4. In the India ink absorption test, India ink granules were not absorbed on the epithelium of the ductus deferens, but the granules reactive to acid phosphatase a, pp.ared in a line on the free border of each parts of the ductus deferens. The granules reactive to alkaline phosphatase were noted on the luminal border of ductus deferens mainly, but weak reaction showed than acid phophatase were a, pp.ared. The granules reactive to PAS were a, pp.ared mostly near on the free border of hte epithelial cells of ductus deferens. 5. Number of sperm, Indes of sperm vitality and MRT in the different parts of ductus deferens were tended to be somewhat dominant in the mid and lower parts than in u, pp.r part, even though not significant in the statistical analysis. Ratio of sperm abnormality was tended to be relatively high in the u, pp.r part too, and in the sperm of abnormality blunted head was less in number significantly in the mid and lower part than in the u, pp.r part.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.497-504
• /
• 2020

Constructing speaker embeddings in speaker verification is an important issue. In general, a self-attention mechanism has been applied for speaker embedding encoding. Previous studies focused on training the self-attention in a high-level layer, such as the last pooling layer. In this case, the effect of low-level layers is not well represented in the speaker embedding encoding. In this study, we propose Masked Cross Self-Attentive Encoding (MCSAE) using ResNet. It focuses on training the features of both high-level and low-level layers. Based on multi-layer aggregation, the output features of each residual layer are used for the MCSAE. In the MCSAE, the interdependence of each input features is trained by cross self-attention module. A random masking regularization module is also applied to prevent overfitting problem. The MCSAE enhances the weight of frames representing the speaker information. Then, the output features are concatenated and encoded in the speaker embedding. Therefore, a more informative speaker embedding is encoded by using the MCSAE. The experimental results showed an equal error rate of 2.63 % using the VoxCeleb1 evaluation dataset. It improved performance compared with the previous self-attentive encoding and state-of-the-art methods.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.414-423
• /
• 2020

In this paper, we propose a Dilated Convolution Gate Linear Unit (DCGLU) to mitigate the lack of sparsity and small receptive field problems caused by the segmentation map extraction process in sound event detection with weak labels. In the advent of deep learning framework, segmentation map extraction approaches have shown improved performance in noisy environments. However, these methods are forced to maintain the size of the feature map to extract the segmentation map as the model would be constructed without a pooling operation. As a result, the performance of these methods is deteriorated with a lack of sparsity and a small receptive field. To mitigate these problems, we utilize GLU to control the flow of information and Dilated Convolutional Neural Networks (DCNNs) to increase the receptive field without additional learning parameters. For the performance evaluation, we employ a URBAN-SED and self-organized bird sound dataset. The relevant experiments show that our proposed DCGLU model outperforms over other baselines. In particular, our method is shown to exhibit robustness against nature sound noises with three Signal to Noise Ratio (SNR) levels (20 dB, 10 dB and 0 dB).

• Proceedings of the Population Association of Korea Conference
• /
• 2006.12a
• /
• pp.113-128
• /
• 2006

The main theme of this paper was to investigate the role of education as a source of economic growth in Korea. In this study, first, the objective mode was built by extending neoclassical Solow growth theory. Second, the capital deepening typical of an endogenous economic per-capita growth model was developed empirically for seven East-Asian economies as for the medium term, during $1975{\sim}2004$. And then we found the meaning of coefficients of growth factors, direct relative contribution of each input to per-capita growth in seven East-Asian countries, relative indirect contribution of education to per-capita growth in Korea, accounting for difference due to accumulation in Korea. The indirect relative contributions of secondary and higher education and R&D to per-capita growth change the results somewhat. Secondary education is still the largest single contributor 83.6 percent of predicted growth is due to secondary school enrollment in Korea. Primary education comes second with 37.5 percent and followed by higher education at -52.9 percent. Physical investment gives 62.3 percent and unimproved raw labor contributes only -1.4 percent.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.15-23
• /
• 2017

The person Re-identification is the most challenging part of computer vision due to the significant changes in human pose and background clutter with occlusions. The picture from non-overlapping cameras enhance the difficulty to distinguish some person from the other. To reach a better performance match, most methods use feature selection and distance metrics separately to get discriminative representations and proper distance to describe the similarity between person and kind of ignoring some significant features. This situation has encouraged us to consider a novel method to deal with this problem. In this paper, we proposed an enhanced recurrent neural network with three-tier hierarchical network for person re-identification. Specifically, the proposed recurrent neural network (RNN) model contain an iterative expectation maximum (EM) algorithm and three-tier Hierarchical network to jointly learn both the discriminative features and metrics distance. The iterative EM algorithm can fully use of the feature extraction ability of convolutional neural network (CNN) which is in series before the RNN. By unsupervised learning, the EM framework can change the labels of the patches and train larger datasets. Through the three-tier hierarchical network, the convolutional neural network, recurrent network and pooling layer can jointly be a feature extractor to better train the network. The experimental result shows that comparing with other researchers' approaches in this field, this method also can get a competitive accuracy. The influence of different component of this method will be analyzed and evaluated in the future research.

• Journal of Environmental Policy
• /
• v.8 no.4
• /
• pp.75-94
• /
• 2009

The process of obtaining third-party financing contacts was analyzed via a two-stage game model: a "signaling game" for the first stage,and a "principal-agent model" for the second stage. The two-stage game was solved by a process of backward induction. In the second stage game, the optimal effort level of the energy saving company (ESCO), the optimal compensation scheme of the energy user, and the optimal payoffs for both parties were derived for each subgame. The optimal solutions forthe different subgames were then compared with each other. Our main finding was that if there is some restriction on ESCO's revenue (e.g. a progressive sales tax) that causes ESCO's revenue toincrease at a decreasing rate, then the optimal sharing ratio is uniquely determined at a level of strictly less than one under a linear compensation scheme, i.e. a unique balance exists. Subgames have a unique equilibrium arrived at separately for each situation,. Within this equilibrium, energy users accept energy audit proposals from H-type ESCOs with high levels of technology, but reject proposals from L-type ESCOs with low levels of technology. While L-type ESCOs cannot attain profits in the third-party financing market, H-type ESCOS can pocket the price differential between L-type and H-type audit fees. Accordingly, revenues in an H-type ESCO equilibrium increase not only in line with the technology of the ESCO inquestion, but also faster than in an L-type equilibrium due to more advanced technology. At the same time, energy users receive some positive payoff by allowing ESCOs to perform third-party financing tasks within their existing energy system without incurring any extra costs.

• The Korean Journal of Air & Space Law and Policy
• /
• v.26 no.1
• /
• pp.93-125
• /
• 2011

The International Civil Aviation Convention (Chicago Convention) has been a backbone of international air transport system whereby air transport between States should be based on bilateral agreements, and in particular, international airfares, which are set up through IATA(International Air Transport Association) rate-fixing machinery could be approved by the governments concerned. International airfares are fares for transporting passenger and freight and their conditions between two or more countries. However, since U.S. pursued th deregulation policy in 1978 whereby routes, capacity and fares could be freely determined by airlines, many States have been following so called open-skies agreements. In many cases, aeronautical and competent authorities have been reviewing whether airlines' commercial activities including air fares could possibly conflict with free competition rules envisaged in relevant laws and regulations. As competition among airlines gets intense, airlines often resort to cooperation with other airlines in the forms such as equity exchange, M&A, code-sharing, fares consultation and resource pooling, mainly with a view to effectively use resources available and to avoid monopoly situation resulting from excessive and destructive competition among players. Whereas bearing in mind that application of competition laws is important to secure consumers' interests by preventing airlines's malpractices such as bargaining exorbitant fares, it is also important to comprehensively consider as many factors as possible, from that unilateral measure by governments may bring about retaliatory measures by the governments affected, to that airlines' cooperative practices may rather increase consumers' benefits by lowering air fares.

• Journal of Korean Society of Forest Science
• /
• v.89 no.2
• /
• pp.167-172
• /
• 2000

Inter-simple sequence repeat (I-SSR) markers were analyzed from diploid genomes of 95 nutmeg trees (Torreya nucifera Siev. et Zucc.) in 5 populations. A total of 62 I-SSR amplicons were observed and 7 of them were monomorphic in 95 individuals. DNA fingerprint of each tree was verified by pooling the observed I-SSR amplicons. Most of the genetic diversity was allocated within population (90.65%) and all the populations revealed similar level of I-SSR amplicon diversity within population. Degree of population differentiation (${\phi}_{ST}=9.35%$) was moderate on the basis of criteria obtained from isozyme analysis. Based on the results of the cluster analysis of UPGMA, genetic relationships among 5 populations were not coincided with the pattern of geographic distribution. Non-significant confidence interval at each node also suggests that all the nutmeg populations are genetically not much differentiated.

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.635-645
• /
• 2007

Component technology has been widely accepted as an effective way for building software systems with reusable components, and Microsoft (MS) .NET is one of the recent representative component technologies. Model Driven Architecture (MDA) is a new development paradigm which generates software by transforming design models automatically and incrementally. Transformation of structural models in MDA has been successfully applied. However, transformation of dynamic models and pervasive services, such as transaction service, security service, synchronization service and object pooling are largely remains as an area for further research. The recent enterprise system has multi tier distributed architecture, and the functionality of early mentioned pervasive services is essential for this architecture. .NET platform can implement Component Object Model+ (COM+) component for supporting pervasive services by specify Attribute code. In this paper, we specify the functionalities of the COM+ component offering pervasive services, and then those functionalities are defined by UML profile. By using the profile, the Platform Specific Model (PSM) for .NET/C# is specified, and .NET components are automatically generated through our tool. The development productivity, extensibility, portability, and maintenance of software can be dramatically improved by using of the proposed methods.

• 한국사회정책
• /
• v.25 no.4
• /
• pp.71-104
• /
• 2018

This paper adopts a distributive performance process model of in-work poverty based on labor markets, households, and welfare states and analyzes the 4-11 waves of the Korean Welfare Panel Study during 2008-15. Previous studies on in-work poverty have focused on the definitions and concepts of in-work poverty by analyzing employment and unemployment persistence and repetition dynamics, but rarely paid attention to institutional distributive performance. In this regard, this study preforms a stepwise analysis of labor markets, households, and welfare states as a process of income generation in labor markets, satisfaction of welfare needs and income pooling at households, and deduction of social security contribution and income tax as well as receipt of public transfer income at welfare states. Results of empirical analysis show that in-work poverty had been on increase during 2008-11, followed by a decrease between 2012-15. At labor market stages, full time status had the most prominent impact on in-work poverty process, while status by employment and contract type have generated a huge variation as well. At household stages, household work intensity and number of earners contributed to reduction of in-work poverty, but the relations did not seen to be straightforward. However, welfare state played little role in lifting employees out of in-work poverty. In terms of institutional distributive process, in-work poverty was prevalent in either household-welfare state stage or labor market-household-welfare stage. Non-vulnerable group in terms of in-risk poverty was around 80% of the sample during the period of analysis, the size of which has remained constant.