• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.4
• /
• pp.371-376
• /
• 2007

This paper presents an emotion recognition and its expression system of an intelligent robot like a home robot or a service robot. Emotion recognition method in the robot is used by a facial image. We use a motion and a position of many facial features. apply a tracking algorithm to recognize a moving user in the mobile robot and eliminate a skin color of a hand and a background without a facial region by using the facial region detecting algorithm in objecting user image. After normalizer operations are the image enlarge or reduction by distance of the detecting facial region and the image revolution transformation by an angel of a face, the mobile robot can object the facial image of a fixing size. And materialize a multi feature selection algorithm to enable robot to recognize an emotion of user. In this paper, used a multi layer perceptron of Artificial Neural Network(ANN) as a pattern recognition art, and a Back Propagation(BP) algorithm as a learning algorithm. Emotion of user that robot recognized is expressed as a graphic LCD. At this time, change two coordinates as the number of times of emotion expressed in ANN, and change a parameter of facial elements(eyes, eyebrows, mouth) as the change of two coordinates. By materializing the system, expressed the complex emotion of human as the avatar of LCD.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.3
• /
• pp.513-521
• /
• 2010

Most of universities have carried out course evaluation to apply the performance appraisal for professor. But, course evaluation depends on characteristics of each class such as class size, type of lecture, evaluator's grade and so on. As the results, such characteristics of each class lead to serious bias which makes lecturers distrust the course evaluation results. Hence, we propose a equating method for the course evaluation by regression analysis which use stepwise variable selection. And we compare proposed method with the other method by Cho et al. (2009) with respect to efficiencies. Also we give the example to which the method is applied.

• Journal of the Korean Society for information Management
• /
• v.35 no.2
• /
• pp.63-88
• /
• 2018

This research evaluated differences of classification performance for feature selection methods using LDA topic model and Doc2Vec which is based on word embedding using deep learning, feature corpus sizes and classification algorithms. In addition to find the feature corpus with high performance of classification, an experiment was conducted using feature corpus was composed differently according to the location of the document and by adjusting the size of the feature corpus. Conclusionally, in the experiments using deep learning evaluate training frequency and specifically considered information for context inference. This study constructed biomedical document dataset, Disease-35083 which consisted biomedical scholarly documents provided by PMC and categorized by the disease category. Throughout the study this research verifies which type and size of feature corpus produces the highest performance and, also suggests some feature corpus which carry an extensibility to specific feature by displaying efficiency during the training time. Additionally, this research compares the differences between deep learning and existing method and suggests an appropriate method by classification environment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.53-60
• /
• 2012

Nearest-neighbor classification predicts the class of an input data with the most frequent class among the near training data of the input data. Even though nearest-neighbor classification doesn't have a training stage, all of the training data are necessary in a predictive stage and the generalization performance depends on the quality of training data. Therefore, as the training data size increase, a nearest-neighbor classification requires the large amount of memory and the large computation time in prediction. In this paper, we propose a prototype selection algorithm that predicts the class of test data with the new set of prototypes which are near-boundary training data. Based on Tomek links and distance metric, the proposed algorithm selects boundary data and decides whether the selected data is added to the set of prototypes by considering classes and distance relationships. In the experiments, the number of prototypes is much smaller than the size of original training data and we takes advantages of storage reduction and fast prediction in a nearest-neighbor classification.

• Journal of the Society of Naval Architects of Korea
• /
• v.55 no.6
• /
• pp.490-496
• /
• 2018

The consistency of the initially designed waves in the domain is essential for accurate calculation of the added resistance in waves through CFD. In particular, unwanted reflected waves at domain boundaries can cause incorrect numerical solutions due to the superposition with initially designed waves. Euler Overlay Method(EOM) is one of the methods for reducing wave reflections by adding an additional source term to momentum and phase conservation equations, respectively. In this study, we apply the Euler Overlay Method(EOM) to the open-source CFD library, OpenFOAM(R), to simulate the accurate free-surface waves in the domain and the parametric study is performed for efficient implementation of Euler Overlay Method(EOM). Considering that the damping efficiency depends on the selection of the overlay parameter in the added source terms, the size of overlay zone and the wave steepness, the influences of these factors are tested through the wave elevation measured at constant time intervals in the 2D numerical wave tank. Through this process, guidelines for selection of optimal overlay parameter and overlay zone size that can be applied according to the scaling law are finally presented.

• Journal of Biosystems Engineering
• /
• v.43 no.4
• /
• pp.296-302
• /
• 2018

Purpose: The aim of this study was to determine appropriate threshing and selection conditions for sorghum harvesting using a rice combine-harvester. Methods: Sorghum harvesting performance was tested using an actual rice combine. Through this test, the grain loss rate and the composition of crops according to the engine and fan speeds of the combine were investigated. Furthermore, the optimal threshing and selection conditions were determined by carrying out a harvest test based on the opening size factor of the concave in a test field. Results: The grain loss rate for the sorghum using a concave ($18{\times}18mm$) of the rice combine was the lowest at 0.1% at a chaffer angle of $40^{\circ}$, engine speed of 2000 rpm, and fan speed of 20 m/s, but the sorting sieve clogged frequently. Furthermore, as the engine speed and fan speed increased, the grain loss rate also increased. The sorghum harvesting test results of the combine according to the concave opening size showed that the grain loss rate was 0.5% at a driving speed of 0.5 m/s, with a concave opening diameter of 13 mm, a chaffer angle of $40^{\circ}$, a concave sieve oscillation frequency of 4.8 Hz, a fan speed of 20 m/s, and an engine speed of 2000 rpm. Conclusions: Findings showed that sorghum could be harvested using a head feeding rice combine.

• Fashion & Textile Research Journal
• /
• v.20 no.4
• /
• pp.464-473
• /
• 2018

The objective of this study is to develop a sizing system of fitness clothing that can properly accommodate various body sizes of Korean senior women. The sizing system of upper and lower fitness clothing was developed in the present study by selection of key variables, identification of size category candidates, and determination of an optimal sizing system. First, key anthropometric dimensions (stature and bust circumference for upper clothing and stature; waist circumference for lower clothing) were identified by factor analysis on the direct body measurements (n = 272) and 3D whole-body scan data (n = 271) of Korean senior women in Size Korea. Second, sizing system candidates based on the key dimensions of upper and lower clothing were explored using a grid method and an optimization method. Lastly, among the sizing system candidates, optimal sizing systems of upper and lower clothing were selected in terms of accommodation rate. Five size categories (short/small, short/medium, tall/small, tall/medium, and tall/large) were selected as the optimal sizing systems of upper and lower clothing with 89% and 78% of accommodation rate, respectively, for the Korean senior women. The anthropometric characteristics of the representative humans of the optimal size categories would be of use in the design of fitness compressive wear for the better fit and effectiveness of exercise and health of Korean senior women.

• Ocean and Polar Research
• /
• v.36 no.2
• /
• pp.199-208
• /
• 2014

Copepods are a major food source for marine fish larvae in nature. Many studies on copepods culture have been conducted to develop a new live food for the seedling production of marine fish larvae. But fish farmers still depend on rotifer and Artemia nauplii. This study was carried out to find suitable copepods as live food for the larvae in hatchery. Eight species of copepods (1 calanoid, 2 cyclpoid, 5 harpacticoid) that were fed Isochrysis galbana were examined in terms of the size of nauplii, fecundity, amino acids, and fatty acids contents. These species were divided into small (nauplii length 46-86 ${\mu}m$) and large (nauplii length 120-188 ${\mu}m$) size group. Nitokra spinipes in the small group and Tigriopus japonicus in the large group showed the highest fecundity with 151.1 and 139.6 nauplii production per gravid female, respectively. With regard to nutrients, essential amino acids were the highest with 21.2% in cyclopoid Paracyclopina nana in the small group and n-3 HUFA were the highest in calanoid Pseudodiaptomus inopinus (8.5 ${\mu}g/mg$) in the large group and P. nana (8.8 ${\mu}g/mg$). In terms of the size, fecundity, and nutritional value of copepods examined in this study, N. spinipes and P. nana seem to be suitable copepod species to develop as a new live food for small mouth fish larvae.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.26 no.2
• /
• pp.293-307
• /
• 1998

This study has been divided into two parts, 1) literature review and 2) a case study. Based on literature review and case studies found in Korea and abroad, this study aimed at classifying wildlife passages and establishing their development procedures in order to establish a theoretical ground for the development of wildlife passage as a way to link habitats that have been fragmented by the construction of roads and other developments. The results of the literature review are as follows : 1) Impact of road construction on surrounding ecological environment is found in various aspects. Along with damage of habitats as direct impact, decrease of biodiversity and simplification of species due to limited breeding and movement are cited as possible indirect impact. 2) In order to lessen the impact generated from fragmented habitats, various types of wildlife passage are being developed in Europe and are showing positive results. 3) In Korea, interests on animal passages have been rising recently. The ministry of Environment and local authorities are establishing development plans for the passages. However, practical materials on the usefulness of adoped methods are not available.4) In order to create an effective wildlife passage, decisions have to be made first on basic factors of planning and designing such as ecosystem rehabilitation methods and surveys on wildlife movement routes and habitat environment. Based on this, in a planning and designing stage, the location and type of a passage, size and measures to attract animals and prevent a collision need to be finalized. This should be followed by a series of process of maintenance, management and monitoring after the development. Based on the results of a theoretical study to develop an wildlife passage, Part 2 of this study deals with the results of an experiment through a specific planning and design that includes a process of ecological surveys, site selection, type selection, ecosystem infrastructure, planting design and construction details on an actual study area and presents them as a model project.

• Asian-Australasian Journal of Animal Sciences
• /
• v.32 no.11
• /
• pp.1657-1663
• /
• 2019

Objective: A genome-based best linear unbiased prediction (GBLUP) method was applied to evaluate accuracies of genomic estimated breeding value (GEBV) of carcass and reproductive traits in Berkshire, Duroc and Yorkshire populations in Korean swine breeding farms. Methods: The data comprised a total of 1,870, 696, and 1,723 genotyped pigs belonging to Berkshire, Duroc and Yorkshire breeds, respectively. Reference populations for carcass traits consisted of 888 Berkshire, 466 Duroc, and 1,208 Yorkshire pigs, and those for reproductive traits comprised 210, 154, and 890 dams for the respective breeds. The carcass traits analyzed were backfat thickness (BFT) and carcass weight (CWT), and the reproductive traits were total number born (TNB) and number born alive (NBA). For each trait, GEBV accuracies were evaluated with a GEBV BLUP model and realized GEBVs. Results: The accuracies under the GBLUP model for BFT and CWT ranged from 0.33-0.72 and 0.33-0.63, respectively. For NBA and TNB, the model accuracies ranged 0.32 to 0.54 and 0.39 to 0.56, respectively. The realized accuracy estimates for BFT and CWT ranged 0.30 to 0.46 and 0.09 to 0.27, respectively, and 0.50 to 0.70 and 0.70 to 0.87 for NBA and TNB, respectively. For the carcass traits, the GEBV accuracies under the GBLUP model were higher than the realized GEBV accuracies across the breed populations, while for reproductive traits the realized accuracies were higher than the model based GEBV accuracies. Conclusion: The genomic prediction accuracy increased with reference population size and heritability of the trait. The GEBV accuracies were also influenced by GEBV estimation method, such that careful selection of animals based on the estimated GEBVs is needed. GEBV accuracy will increase with a larger sized reference population, which would be more beneficial for traits with low heritability such as reproductive traits.