• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.562-567
• /
• 2009

As a key mechanism of the human emotion interaction, Facial Expression is a powerful tools in HRI(Human Robot Interface) such as Human Computer Interface. By using a facial expression, we can bring out various reaction correspond to emotional state of user in HCI(Human Computer Interaction). Also it can infer that suitable services to supply user from service agents such as intelligent robot. In this article, We addresses the issue of expressive face modeling using an advanced active appearance model for facial emotion recognition. We consider the six universal emotional categories that are defined by Ekman. In human face, emotions are most widely represented with eyes and mouth expression. If we want to recognize the human's emotion from this facial image, we need to extract feature points such as Action Unit(AU) of Ekman. Active Appearance Model (AAM) is one of the commonly used methods for facial feature extraction and it can be applied to construct AU. Regarding the traditional AAM depends on the setting of the initial parameters of the model and this paper introduces a facial emotion recognizing method based on which is combined Advanced AAM with Bayesian Network. Firstly, we obtain the reconstructive parameters of the new gray-scale image by sample-based learning and use them to reconstruct the shape and texture of the new image and calculate the initial parameters of the AAM by the reconstructed facial model. Then reduce the distance error between the model and the target contour by adjusting the parameters of the model. Finally get the model which is matched with the facial feature outline after several iterations and use them to recognize the facial emotion by using Bayesian Network.

• Journal of the Korean School Mathematics Society
• /
• v.22 no.4
• /
• pp.461-482
• /
• 2019

The purpose of this study was to analyze the elementary pre-service teachers' perspectives on the use of the elementary mathematics textbooks by observing mathematics classes at the schools and to present suggestions on the use of textbooks in the mathematics classes. For this study, let 54 pre-service elementary school teachers observe a total of 54 classes (1,299 students) in the first through sixth grades at 15 areas in Seoul city. With the observations of the mathematics classes, the researcher analyzed the pre-service elementary teachers' perspectives on the use of resources including mathematics textbooks in the elementary mathematics classes. As a result of the study, the pre-service teachers recognized that mathematics textbooks are the main material for mathematics classes, but they think that they need to reconstruct the contents of the textbooks for effective teaching and learning mathematics. In addition, various additional materials need to be used in order to help students learn mathematics more effectively, including by differentiated materials. Also, instruments used in the classes were mainly consisted of TV and PPT data. Pre-service teachers were not much satisfied with the contents of the elementary mathematics textbooks. However, they did not feel the need for development of digital textbooks. The researcher suggested a long term research on how to use study materials such as textbooks within the overall mechanism of teaching and learning processes. Also, practical teacher education programs that include applications of textbooks at schools are necessary in order to improve the quality of mathematics teaching and learning.

• Journal of Korean Philosophical Society
• /
• v.116
• /
• pp.257-280
• /
• 2010

I try to formalize the system of modal logic and interpret it in view of constructivism through this study. As to the meaning of a sentence, as we saw, Frege endorsed extensions in view of the fact that they are enough to provide for a compositional account for truth, in particular that (1) the assignment of extensions to expressions is compositional ; (2) the assignment of extensions to sentences coincides with the assignment of truth values. But nobody would be willing to admit that a truth value is what a sentence means and that consequently all true sentences are synonymous. So, if what we are after is meaning in the intuitive sense, then extensions would not do. This consideration has later become the point of departure of modal and intensional semantics. So, it is clear that the language of modal logic do not allow for an extensional interpretation. ${\square}$ is syntactically on a par with ${\vdash}$, hence within the extensional framework it would have to denote a unary truth function. This means that if modal logic is to be interpreted, we need a semantics which is not extensional. The first attempt to build a feasible intensional semantics was presented by Saul Kripke. He came to the conclusion that we must let sentences denote not truth values, but rather subsets of a given set. He called elements of the underlying set possible world. Hence each sentence is taken to denote the set of those possible world in which it is true. This lets us explicate necessity as 'truth in every possible world' and possibility as 'truth in at least one possible world'. But it is clear that the system of modal logic is not only an enlargement of propositional logic, as long as the former contains the new symbols, but that it is of an other nature. In fact, the modal logic is intensional, in that the operators do not determine the functions of truth any more. But this new element is not given a priori, but a posteriori from construction by logicist.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.1
• /
• pp.58-63
• /
• 2022

This study was conducted to examine the characteristics of pitch perception and temporal resolution through Within-/Cross-Channel Gap Detection Thresholds (WC/CC GDTs) using low-frequency pure tones (such as 264 Hz, 373 Hz and 528 Hz related to C4, C4#, and C5 musical tones. 40 young people and 20 elderly people with normal hearing participated in this study. The results of WC GDTs were approximately 2 ms ~ 4 ms threshold values regardless of frequencies in two groups. There was no statistically significant difference in WC GDTs between groups. In both groups, CC GDTs were larger than WC GDTs, and as the frequency difference increased, the CC GDTs also increased. In particular, in the comparison between groups of CC GDTs, the results of the elderly group were 8 times ~ 10 times larger than that of the young group, and there was a statistically significant difference between the groups. These data also showed a different trend of GDTs in comparison with the previous data obtained from musical stimuli.This study suggests that GDTs may influence pitch perception mechanisms and can be used as psychoacoustic evidence for nonlinear responses of auditory nervous system.

• Journal of the Korea Institute of Building Construction
• /
• v.12 no.1
• /
• pp.122-132
• /
• 2012

Establishing a success model of a specific information system is critical to understanding the mechanism of IS success, the various dimensions of IS performance, and the factors and their causal relations in IS success. As one of the key IT applications, the project management information system (PMIS), particularly the web-based PMIS (Web-PMIS), has played a significant role in construction management processes in Korea. However, there have been few research attempts made to construct a Web-PMIS success model. This study primarily aims to propose a Web-PMIS success model based on DeLone and McLean's IS success model, and to discuss whether or not the D&M IS success model can be applied to the construction Web-PMIS. A questionnaire was sent out to Web-PMIS users (construction managers and constructors), and 253 completed questionnaires were received. Through multi-regression analysis, it was confirmed that it is statistically acceptable to apply the D&M IS success model to the Web-PMIS. However, the explanatory power of the model is not sufficient, and some of the model factors are not statistically significant enough. Relying on the statistical analysis results, this study also discusses the development direction for the Web-PMIS success model.

• Korean Journal of Cognitive Science
• /
• v.17 no.4
• /
• pp.357-373
• /
• 2006

Virtual reality (VR) provides a virtual experiment (VE) context consisting of information presented to the senses of the user. The user perceiver and interprets the VE context, and then naturally recognizes a level of realism in the VE. Presence is often thought of as the sense of 'being there' in the n. Presence includes overall feelings about the information conveyed from a virtual avatar to the user. Therefore, there must be brain mechanisms for integrating sensory information about presence.'Feeling of presence' is related with the user's cognition and perception about information on communication through medium. Thus 'feeling of presence' may characterize perceptual mechanisms in the brain. We studied these mechanisms by presenting a VR that consisted of an avatar telling a story about a social conversation. We performed covariance analysis on subjective brain activity (fMRI) during the story presentation with a presence score. The data analysis revealed that activity in several brain areas was correlated with the presence store. A positive correlation was shown in the right lingual gyrus, right cuneus, left lingual gyrus, right fusiform gyrus, left inferior temporal gyrus, anterior cingulate cortex and right posterior cingulate cortex of the brain. This study showed the brain mechanism to be related the feeling of presence and brain activities in our subjects, using VR to communicate information.

• Journal of Technology Innovation
• /
• v.22 no.3
• /
• pp.225-260
• /
• 2014

In this paper, we present a literature review of the mechanisms and methodology for technology spillover, in order to provide a theoretical and practical foundation for estimating the technology spillover effect of public funded R&D programs. Our research consists of four parts. First, we investigate what technology spillover is and how it works. Second, we review the methodology used to empirically measure technology spillover, with a view to understanding the advantages and disadvantages of each method. Third, we identify the gap between the spillover phenomenon and its estimated results, caused by the limitations of the methodology, and discuss the challenges in measuring technology spillover. Finally, we present several considerations to improve the measurement of technology spillover in the context of public funded R&D programs. Our paper provides policy makers and researchers with basic knowledge about technology spillover and helps them suitably evaluate the results obtained from the literature review. In addition, it contributes towards improving the existing methodology by recognizing the gap between the spillover phenomenon and the methodology.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.40 no.12
• /
• pp.1067-1076
• /
• 2016

This paper describes the development of a hybrid-powered wheelchair (HPW) for the elderly. The proposed HPW has novel mechanical and control features compared with conventional powered wheelchairs. An ergonomic back-braking mechanism was designed in order to stop the wheels easily. In terms of control features, the HPW remarkably reduces the muscle power required by combining various assistive functions, such as wheel torque assistance, friction/inertia compensation, gravity compensation, and the one-hand driving algorithm. For wheel torque assistance, strain gauges were attached to the hand-rim in order to measure the wheel torque applied by a human. Gyroscopes and an accelerometer were attached to the wheel and chair respectively for friction and inertia compensation. An inclinometer was attached for gravity compensation and the one-hand driving algorithm was included for patients who can only use one hand. The one-hand driving algorithm controls the angular velocity of the uncontrolled wheel by using a gyroscope and pressure sensors attached to the bottom of the seat. Finally, the performance of the proposed motion assisted algorithm was verified through various experiments.

• The Journal of the Korea Contents Association
• /
• v.22 no.9
• /
• pp.95-103
• /
• 2022

This study aims to understand a gamer participatory mechanism by comparing Gamer Participation 1.0(Bartz Liberation War) and Gamer Participation 2.0(Truck Demonstration). The process of gamers' political participation is analyzed by applying the communication-mediated O-S-O-R model with the case of gamers' truck protests caused by probability items' issue. This study found out that changes in the social, technological, and discursive layers that constitute the gamer's communication ecology led to diversification of communication channels and changes in the perception of games. Gamers utilized the technological layer of the media environment that presupposes immediate mutual communication, expressed opinions on issues in the mobilization process, and shared the necessity of participation to derive collective mobilization. Through communication, gamers were able to participate in socio-political issues with high participation thresholds. This study is significant in that it discussed the gamers' democratic citizenship and role as issue publics. The study suggests that the need for theoretical and methodological expansion to analyze various participatory cases.