• The Journal of the Korea Contents Association
• /
• v.20 no.2
• /
• pp.594-606
• /
• 2020

On the contrary to the steady rise of social problems about depression, only a few take care of mental health themselves due to their negative perception of psychotherapy. In this study, we propose service design and user interface design of mobile applications that introduce an automatic emotional recognition system to prevent leaving depression unattended. First, we have come up with improvements in the usability of existing apps through literature research and case analysis, and draw out Personas from surveys and in-depth interviews of young people. As an essential element of the design of the Depression Improvement Service, we recommend having three components: writing a diary to manage schedule, recognize users' emotional state through avatar management, and tracking down emotional points to improve the cognitive system. Through this proposal, we want to prevent the development of depression and help to resolve the changes in perception of depression and longitude depression on our own. Ultimately, we publicize the necessity of mental health care application design and expect to lower the personal and social costs for psychiatric care.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.2
• /
• pp.85-89
• /
• 1996

In this paper, transition constrained Hidden Markov Model(HMM) in which the transition between states occur only within prescribed time slot is proposed and the performance is evaluated in the noisy environment. The transition constrained HMM can explicitly limit the state durations and accurately de scribe the temporal structure of speech signal simply and efficiently. The transition constrained HMM is not only superior to the conventional HMM but also require much less computation time. In order to evaluate the performance of the transition constrained HMM, speaker independent isolated word recognition experiments were conducted using semi-continuous HMM with the noisy speech for 20, 10, 0 dB SNR. Experiment results show that the proposed method is robust to the environmental noise. The 81.08% and 75.36% word recognition rates for conventional HMM was increased by 7.31% and 10.35%, respectively, by using transition constrained HMM when two kinds of noises are added with 10dB SNR.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.8
• /
• pp.5565-5570
• /
• 2015

Implicit citation sentence recognition is to locate citation sentences which lacks explicit citation markers, from articles' full-text. State-of-the-art approaches exploit word ngrams, clue words, researcher's surnames, mentions of previous methods, and distance relative to nearest explicit citation sentences, etc., reaching over 50% performance. However, most previous works have been conducted on English. As for Korean, a rule-based method using positive/negative clue patterns was reported to attain the performance of 42%, requiring further improvement. This study attempted to learn to recognize implicit citation sentences from Korean literatures' full-text using Korean lexical features. Different lexical feature units such as Eojeol, morpheme, and Eumjeol were evaluated to determine proper lexical features for Korean implicit citation sentence recognition. In addition, lexical features were combined with the position features representing backward/forward proximities to explicit citation sentences, improving the performance up to over 50%.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.63-71
• /
• 2008

In this paper, we propose a color based hierarchically propagated particle filter that extends the color based particle filter into the articulated upper body tracking. Since color feature is robust to partial occlusion and rotation, the color based particle filter is widely used for object tracking. However, in articulated body tacking, it is not desirable to use the traditional particle filter because the dimension of the state vector usually is high and thus, many samples are required for robust hacking. To overcome this problem, we use a hierarchical tracking method for each body part based on the blown body part. By using a hierarchical tracking method, we can reduce the number of samples for robust tracking in the cluttered environment. Also for human pose recognition, we classify the human pose into eight categories using Support Vector Machine(SVM) according to the angle between upper- arm and fore-arm. Experimental results show that our proposed method is more efficient than the traditional particle filter.

• Journal of The Korean Association For Science Education
• /
• v.33 no.5
• /
• pp.981-994
• /
• 2013

This study aims to examine two hypothetical models for the variables of students' environmental perspective. First hypothetical model is the mediating role of recognition and familiarity between aesthetic and negativistic perspectives. Second hypothetical model is the separate path from humanistic to dominionistic perspectives. One hundred four tenth grade students participated in this study. We used Pearson correlation, partial correlation, and path analysis to examine the fitness of hypothetical models. The findings showed that the mediating role of recognition and familiarity between aesthetic and negativistic perspectives were statistically accepted. To prevent students' bias for fancy or gross animals, the learning (for instance, recognition and familiarity) may play role in reducing the bias. Second, there were two differential paths from humanistic to dominionistic perspectives (ecologistic-naturalistic path and scientistic-utilitarian path). While ecologistic-naturalistic path does not reach dominionistic perspective, scientistic-utilitarian path does reach dominionistic perspective. To prevent students' dominionistic perspective for nature, they need to understand ecologistic-naturalistic minds for the nature.

• Journal of the Ergonomics Society of Korea
• /
• v.28 no.3
• /
• pp.63-71
• /
• 2009

A car navigation system as an in-vehicle route guidance information (RGI) offers a state-of-the-art technological solution to driver navigation in an unfamiliar area. However, the RGI is provided by some pre-determined options in terms of the interface between a driver and a car navigation system. Drivers occasionally pass the target intersection owing to non- or late- recognizing it. This paper is examined the position of driver's turn signal operation and intersection recognition approaching at the target intersection which is difficult to identify, as a fundamental research on developing the additional RGI connecting with the turn signal control. The field experiment was conducted to measure distances of the turn signal operation and the intersection recognition from the target intersection according to left turns, right turns, and landmarks at adjacent intersection. And glance behavior to the car navigation display was evaluated by using an eye camera. The results of the field study indicate that, most case of driving, drivers operate the turn signal until 40m to 50m before coming to the target intersection. The driving simulator experiment was performed to examine the effectiveness of providing the additional RGI when drivers did not operate the turn signal approaching at the target intersection based on the results of the field study. To provide the additional RGI is effective for the intersection identification and recognition, and expected to improve the traffic safety and the comfort for drivers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.8
• /
• pp.1941-1947
• /
• 2014

Document image binarization is largely used as previous stage of document recognition. And the result of document recognition is much affected from the result of document image binarization. There were many studies to binarize document images. The results of previous studies for document image binarization is varied according to the state of document images. In this paper, we propose a technique for document image binarization using MSER that is applied to extract objects from an image. At first, raw MSER objects are extracted from a document image. Because the raw MSER objects cannot be used for document image binarization, the extracted raw MSER objects are modified. Then the final MSER objects are used for document image binarization with the contrast image that is extracted from the document image. Experimental results show that the proposed technique is useful for document image binarization.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1375-1385
• /
• 2022

In this paper, we propose a semi-supervised domain adaptation solution to deal with practical face recognition (FR) scenarios where a single face image for each target identity (to be recognized) is only available in the training phase. Main goal of the proposed method is to reduce the discrepancy between the target and the source domain face images, which ultimately improves FR performances. The proposed method is based on the Domain Adatation network (DAN) using an MMD loss function to reduce the discrepancy between domains. In order to train more effectively, we develop a novel loss function learning strategy in which MMD loss and cross-entropy loss functions are adopted by using different weights according to the progress of each epoch during the learning. The proposed weight adoptation focuses on the training of the source domain in the initial learning phase to learn facial feature information such as eyes, nose, and mouth. After the initial learning is completed, the resulting feature information is used to training a deep network using the target domain images. To evaluate the effectiveness of the proposed method, FR performances were evaluated with pretrained model trained only with CASIA-webface (source images) and fine-tuned model trained only with FERET's gallery (target images) under the same FR scenarios. The experimental results showed that the proposed semi-supervised domain adaptation can be improved by 24.78% compared to the pre-trained model and 28.42% compared to the fine-tuned model. In addition, the proposed method outperformed other state-of-the-arts domain adaptation approaches by 9.41%.

• The Korean Journal of Community Living Science
• /
• v.15 no.3
• /
• pp.137-148
• /
• 2004

The purpose of this study is to prevent and review counter-action for the sexual harassment in campus by revealing current state of sexual harassment in campus relatively weak in sexual discrimination with regard to sexual harassment emerged as social problem serious in our nation. The targets of study were students and assistants in three schools located at $Kwangju\cdot{Chonnam}$ area, a questionnaire made by pre-researcher was applied to student vs. assistant with regard to sexual consciousness and recognition, and current state of sexual harassment, with analysis of material processed through statistical analysis using SPSSWIN 10.0 for verification of $$\chi$^2$ for identification of difference vs. family and student vs. assistant. The result of this research with regard to sexual harassment is under following: Firstly, the recognition of concept of sexual harassment was proved to be aware of be more than half a responders above average, lack of recognition showed significant number requiring establishment of concept of sexual harassment, the responders stated that they could do it alone should be treated in the counseling room in school. Secondly, 28.7% of students and assistants answered that they have suffered from sexual harassment now and. before, even a great number of males have experienced such sexual harassment. In case of female assistants, more than half of them suffered from sexual harassment suggesting us significant level of sexual harassment to female assistants. A depth of sexual harassment to assistants should be investigated in a further way by preparing for and establishing concrete preventive actions and researches, mostly suffered by males; In this respect, a preventive training of sexual harassment should be deemed to be provided to males. Thirdly, as a result of investigation of sex education and preventive action against sexual harassment in college, most of elementary, junior, junior high school hardly provide no or inefficient sexual harassment, and they responded that the campus needs its education. As a way to educate campus with sexual harassment efficiently, a special lecture or regular class might be designed for systematic education or sexual harassment through schools counseling room might be effective emphasizing upon role of the room according to opinions of responders. The college is required to install such offices or centers as can exclusively manage and take charge of sex education program operation, legal system of anti-sexual harassment, or personnel arrangement for such activities in school by reflecting such result into school for appropriate and efficient education of or counseling to sex.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.139-147
• /
• 2015

Steady state visual evoked potential (SSVEP) has been actively studied because of its short training time, relatively higher signal-to-noise ratio, and higher information transfer rate. There are two popular analysis methods for SSVEP signals: power spectral density analysis (PSDA) and canonical correlation analysis (CCA). However, the PSDA is known to be vulnerable to noise due to the use of a single channel. Although conventional CCA is more accurate than PSDA, it may not be appropriate for the real-time SSVEP-based BCI system when it has short time window length because it uses sinusoidal signals as references. Therefore, the two methods are not efficient for the real-time BCI system that requires a short TW and a high recognition accuracy. To overcome this limitation of the conventional methods, this paper proposes a frequency recognition method with a combination of CCA and PSDA using the difference between powers of canonical variables obtained from the results of CCA. Experimental results show that the performance of the combination of CCA and PSDA is better than that of CCA for the case of a short TW.