• Proceedings of the Korea Contents Association Conference
• /
• 2006.05a
• /
• pp.273-278
• /
• 2006

In the virtual space in which human imagination exists, a great deal of games and animation contents have been developed and shown. Games and animations are performing a role as an unrealistic representative, and creating abundant virtual culture with variety of human life. A representative, that is, a main character is being completed with external design, unique personality and ability, and action in the story. This study is expecting that a game character might develop into motion acting for emotional situation from the simple action. As giving a game character a role as a medium to express various circumstances and emotions, choosing motions from a pantomimist and applying them to 3D character, and expressed according to the motion acting. Motion acting is extended from the basic motion into the emotional phases such as joy, anger, sorrow and pleasure. The motion acting expressed from various viewpoints is suggested, as communicating clear message through emotional acting contrary to the symbolic language.

• The KIPS Transactions:PartB
• /
• v.9B no.4
• /
• pp.445-452
• /
• 2002

In Korean-English mixed documents, translated technical words are usually used with the attached full words or original words enclosed with parenthesis. In this paper, a collective method is presented to recognize and extract the translation phrases with using a base translation dictionary. In order to process the unregistered title words and translation words in the dictionary, a phonetic similarity matching method, a translation partial matching method, and a compound word matching method are newly proposed. The experiment result of each method was measured in F-measure(the alpha is set to 0.4) ; exact matching of dictionary terms as a baseline method showed 23.8%, the hybrid method of translation partial matching and phonetic similarity matching 75.9%, and the compound word matching method including the hybrid method 77.3%, which is 3.25 times better than the baseline method.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.14 no.2
• /
• pp.57-65
• /
• 2010

To determine the static buckling loads and evaluate the structural performance of slender steel pipe-arches such as for greenhouse structures, a series of modal tests using a fixed hammer and roving sensors was carried out, by providing no load, then a range of vertical loads, on an arch rib in several steps. More attention was given to an internal arch where vertical and horizontal auxiliary members are not placed, unlike an end arch. Modal parameters such as natural frequencies, mode shapes and damping ratios were extracted using more advanced system identification methods such as PolyMAX (Polyreference Least-Squares Complex Frequency Domain), and compared with those predicted by commercial FEA (Finite Element Analysis) software ANSYS for various conditions. A good correlation between them was achieved in an overall sense, however the reduction of natural frequencies due to the existence of preaxial loads was not apparent when the vertical load level was about up to 38% of its resistance. Some difficulties related to the field testing and parameter extraction for a very slender arch, as might arise from the influences of neighboring members, are carefully discussed.

• Journal of the Korean Geographical Society
• /
• v.46 no.1
• /
• pp.20-35
• /
• 2011

This study suggested the systematic steps for quality control, construction of the climatological reference series and homogeneity test and adjustment of temperature series in the Republic of Korea. It also attempted to evaluate more accurate magnitude of change using adjusted temperature data. All erroneous values produced by quality control were detected by internal inconsistency check. The method selected for homogeneity test in this study well defined fairly correct signals of station relocations. Therefore, this method might be regarded as the appropriate one to test homogeneity of temperature series of the Republic of Korea. The increase of temperature of the Republic of Korea after the adjustment were bigger than before the adjustment of annual and seasonal mean temperature. Adjusted temperature data produced by these steps will enable to evaluate more accurate characteristics and magnitude of climate change.

• Journal of IKEEE
• /
• v.14 no.3
• /
• pp.224-231
• /
• 2010

Automatic detection of objects in images has been one of core challenges in the areas such as computer vision and pattern analysis. Especially, with the recent deployment of personal mobile devices such as smart phone, such technology is required to be transported to them. Usually, these smart phone users are equipped with devices such as camera, GPS, and gyroscope and provide various services through user-friendly interface. However, the smart phones fail to give excellent performance due to limited system resources. In this paper, we propose a new scheme to improve object recognition performance based on pre-computation and simple local features. In the pre-processing, we first find several representative parts from similar type objects and classify them. In addition, we extract features from each classified part and train them using regression functions. For a given query image, we first find candidate representative parts and compare them with trained information to recognize objects. Through experiments, we have shown that our proposed scheme can achieve resonable performance.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.4
• /
• pp.256-261
• /
• 2018

There are three types of noise generated inside the vehicle: BSR (Buzz, Squeak, Rattle). In this paper, we propose a classifier that automatically classifies automotive BSR noise by using features extracted from deep convolutional neural networks. In the preprocessing process, the features of above three noises are represented as noise-map using STFT (Short-time Fourier Transform) algorithm. In order to cope with the problem that the position of the actual noise is unknown in the part of the generated noise map, the noise map is divided using the sliding window method. In this paper, internal parameter of the deep convolutional neural networks is visualized using the t-SNE (t-Stochastic Neighbor Embedding) algorithm, and the misclassified data is analyzed in a qualitative way. In order to analyze the classified data, the similarity of the noise type was quantified by SSIM (Structural Similarity Index) value, and it was found that the retractor tremble sound is most similar to the normal travel sound. The classifier of the proposed method compared with other classifiers of machine learning method recorded the highest classification accuracy (99.15 %).

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.116-122
• /
• 2010

As the importance of HCI(Human-Computer Interface) caused by dramatically developed computer technology is getting high, the requirement for the design of human friendly systems is also getting high. First of all, the personality and Emotional factor should be considered for implementing more human friendly systems. Many studies on Knowledge, Emotion and personality have been made, but the combined methods connecting these three factors is not so many investigated yet. It is known that memorizing process includes not only knowledge but also the emotion and the emotion state has much effects on the process of reasoning and decision making step. Accordingly, for implementing more human friendly efficient sophisticated intelligent system, the system considering these three factors should be modeled and designed. In this paper, knowledge-emotion reaction model was designed. Five types are defined for representing the personality and emotion reaction mechanism calculating emotion vector based on the extracted Thought threads by Type matching selection was proposed. This system is applied to the virtual memory and its emotional reactions are simulated.

• Science of Emotion and Sensibility
• /
• v.16 no.1
• /
• pp.117-124
• /
• 2013

This study examined that how precisely MFCC, LPC, energy, and pitch related parameters of the speech data, which have been used mainly for voice recognition system could predict the vocal emotion categories as well as dimensions of vocal emotion. 110 college students participated in this experiment. For more realistic emotional response, we used well defined emotion-inducing stimuli. This study analyzed the relationship between the parameters of MFCC, LPC, energy, and pitch of the speech data and four emotional dimensions (valence, arousal, intensity, and potency). Because dimensional approach is more useful for realistic emotion classification. It results in the best vocal cue parameters for predicting each of dimensions by stepwise multiple regression analysis. Emotion categorizing accuracy analyzed by LDA is 62.7%, and four dimension regression models are statistically significant, p<.001. Consequently, this result showed the possibility that the parameters could also be applied to spontaneous vocal emotion recognition.

• Journal of The Korean Association For Science Education
• /
• v.32 no.2
• /
• pp.346-371
• /
• 2012

The purpose of this study is to investigate the possibilities of science ethics education with history of science (HOS) and to develop its teaching and learning model for secondary school students. A total of 72 cases about science ethics were extracted from 20 or more HOS books, journal articles, and newspaper articles. These cases were categorized into 8 areas, such as forgery, fabrication, violation of bioethics in testing, plagiarism and stealth, unfair allocation of credit, over slander, conjunction with ideologies, and social responsibility problems. The results of this study are as follows. First, research forgery, occurring in the process of the research, was the most frequent in HOS. Second, we developed eight teaching lesson plans for each area. Third, we proposed a teaching and learning model based on the developed lesson plans as well as related teaching and learning models in the fields of science ethics education, ethics education, and history education. Our model has five steps, 'investigating-suggesting casesclarifying problems-finding alternatives-summarizing'.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.9
• /
• pp.1165-1171
• /
• 2018

CNN requires a large amount of computation and memory in the process of extracting the feature of the object. Also, It is trained from the network that the user has configured, and because the structure of the network is fixed, it can not be modified during training and it is also difficult to use it in a mobile device with low computing power. To solve these problems, we apply a pruning method to the pre-trained weight file to reduce computation and memory requirements. This method consists of three steps. First, all the weights of the pre-trained network file are retrieved for each layer. Second, take an absolute value for the weight of each layer and obtain the average. After setting the average to a threshold, remove the weight below the threshold. Finally, the network file applied the pruning method is re-trained. We experimented with LeNet-5 and AlexNet, achieved 31x on LeNet-5 and 12x on AlexNet.