• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2023.11a
• /
• pp.202-203
• /
• 2023

최근 화재 탐지 분야는 불꽃 연기의 특징과 인공지능 인식(Detection) 모델을 활용하여 탐지율을 높이려는 연구가 많이 진행되어 왔다. 기존 화재 탐지 정확도를 높이기 위한 모델 연구 이외에도 불꽃·연기의 특징을 다양한 방법으로 데이터 가공한 학습 데이터셋을 활용하는 연구들이 진행되고 있다. 본 논문에서는 화재 탐지시 불꽃/연기의 오탐지율이 높은 것을 확인하고 오탐지율을 낮추기 위해 화재 상황을 인식하여 분류하는 방법과 데이터셋을 제안한다. 제안한 모델은 동영상을 학습데이터로 활용하여 화재 상황의 특징을 추출하여 분류모델에 적용하였다. 평가는 한국정보화진흥원(NIA)에서 진행하는 화재 데이터셋을 이용하여 Yolov8, Slowfast의 모델 성능을 비교 및 분석하였다.

• 한국HCI학회:학술대회논문집
• /
• 2006.02b
• /
• pp.396-402
• /
• 2006

최근 청소년층의 인터넷 이용이 급격히 증가하면서, 초등학생 중 90% 이상이 인터넷을 이용하고 있으며, 그 연령대도 낮아져 저학년 어린이의 이용률이 지속적으로 높아지고 있다. 또한 청소년의 30%가 인터넷 중독 또는 중독 직전에 있는 것이 실태이기도 하다. 10대가 현재 가장 많이 사용하고 있는 정보매체인 컴퓨터는, 그것을 사용함에 있어 교육적으로 도움이 되는 면도 있지만 반면에 무분별한 인터넷 용어나 비속어의 사용, 무절제한 지속적인 이용, 유해사이트에의 노출 등 많은 문제가 산재해 있다. 기존의 유해사이트 차단과 컴퓨터 사용시간 제한 등의 사용 관리를 위한 프로그램들이 많이 선보여진 상태이지만 그 활용 면에서 보면, 좀 더 사용자의 특성을 이해하여 어른들보다 절제가 어려운 어린이 사용자를 위해 부모 혹은 선생님과 같은 관리자와 가르침의 역할을 해줄 수 있는 부분이 더욱 필요하다. 본 연구에서는 무형의 소프트웨어 프로그램이 사용자에게 명령을 내리고 그것을 수행하게 하는 것 보다, 사용자가 친숙하게 받아들이고 정서적으로 접근할 수 있는 사용자와 인터렉션이 가능한 캐릭터 형태의 물리적 아바타를 제안하고자 한다. 또한 비속어나 그와 비슷한 단어들을 사용할 시 의미적으로 비속어인지 아닌지 아바타가 추론할 수 있는 기능으로, 인공지능을 부여하여 사용자가 인터넷 채팅이나 게임을 할때 아바타로 하여금 사용자와의 인터렉션을 통해서 올바르지 않은 언어사용에 대한 인지를 쉽게 하고 개선의 효과를 높이며 지속적인 관심을 이끌 수 있는 인공지능을 가진 캐릭터 아바타를 제안하고자 한다.

• Journal of Korean Ophthalmic Optics Society
• /
• v.20 no.1
• /
• pp.9-14
• /
• 2015

Purpose: In this paper, the changes in eyeglasses frames were investigated when the frames were exposed to artificial sweat. Methods: The changes in surface, hardness, and tensile strength of eyeglasses frames made of nickel silver, Monel, and titanium were tested after the exposure to the artificial sweat of pH 6.5, 6.0, and 5.0, during 1, 3, and 6 week(s), for each. Results: Corrosion pattern of eyeglasses frames made of nickel silver, Monel, and titanium was induced by artificial sweat, and the patterns were different from another frame materials. In proportion to the pH of artificial sweat and exposure time to artificial sweat, hardness of all frames made of three different material decreased, however, tensile strength of the three frames was not changed. Conclusions: The necessity of routine management for eyeglasses frames can be suggested since the changes in surface corrosion and hardness of eyeglasses frames induced by sweat were shown.

• Journal of Digital Convergence
• /
• v.14 no.11
• /
• pp.369-377
• /
• 2016

This study examined artificial joint replacement surgery for early intervention and prevention of the occurrence of delirium surgery. Data of study were analysed using the sample through the EMR (Electronic Medical Record) and after surgery to provide basic data. The subjects were elderly aged 60 years or more and the number of the sample was 821. Data were analysed by using SPSS 20.0 with t-test, $x^2$-test and multiple logistic regression analysis. The study results showed patients with artificial joint replacement surgery incidence of delirium was 13.5%, findings of these variables insisted that the main influencing factors of delirium were caused by age, fall history, physical activity, emotional status, body mass index (BMI) before surgery. The study suggested that the above findings are required for early intervention, early detection and prevention of delirium.

• Journal of the Society of Disaster Information
• /
• v.19 no.4
• /
• pp.968-975
• /
• 2023

Purpose: We would like to confirm that the false positive rate of flames/smoke is high when detecting fires. Propose a method and dataset to recognize and classify fire situations to reduce the false detection rate. Method: Using the video as learning data, the characteristics of the fire situation were extracted and applied to the classification model. For evaluation, the model performance of Yolov8 and Slowfast were compared and analyzed using the fire dataset conducted by the National Information Society Agency (NIA). Result: YOLO's detection performance varies sensitively depending on the influence of the background, and it was unable to properly detect fires even when the fire scale was too large or too small. Since SlowFast learns the time axis of the video, we confirmed that detects fire excellently even in situations where the shape of an atypical object cannot be clearly inferred because the surrounding area is blurry or bright. Conclusion: It was confirmed that the fire detection rate was more appropriate when using a video-based artificial intelligence detection model rather than using image data.

• Journal of Music and Human Behavior
• /
• v.15 no.2
• /
• pp.69-92
• /
• 2018

The purpose of this study was to investigate whether a songwriting intervention would improve self-expression of adolescents with cochlear implants (CI). In this study, the songwriting intervention was designed to incorporate previous research findings on music perception of individuals with CIs and to provide musical experiences in which adolescents with CIs could be successful in using music resources for their self-expression. Accordingly, stepwise songwriting procedures were constructed and music technology (i.e., GarageBand as an iPad application) was used. Two adolescents with CIs participated in the intervention once a week for 8 weeks. Changes in level of participation in songwriting tasks, verbal expression, self-reports about the songwriting process were examined. The results demonstrated that the participants showed increased attempts to express themselves via music making. It was also shown that their negative perception of their ability to verbally express themselves decreased. They were observed as becoming more competent in creating music by deciding how various musical elements should be played and making rhythm or melodic patterns by themselves. These results indicate that the songwriting intervention can facilitate self-expression of adolescents with CIs. Further implications in terms of the psychosocial needs of individuals with CIs are presented.

• Proceedings of the KOSOMBE Conference
• /
• v.1994 no.12
• /
• pp.161-164
• /
• 1994

• Proceedings of the KOSOMBE Conference
• /
• v.1993 no.05
• /
• pp.102-104
• /
• 1993

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.508-510
• /
• 2019

본 연구에서는 딥 러닝을 이용하여 완전 자동화된 다중 클래스 전체 심장 분할 알고리즘을 제안하였다. 제안된 방법은 recurrent convolutional block과 residual multi-dilated block을 삽입하여 기존 U-Net을 개선한 인공신경망 모델을 사용하였다. 평가는 자동화 분석 결과와 수동 평가를 비교하였다. 그 결과 96.88%의 평균 DSC, 95.60%의 정확도, 97.00%의 recall을 얻었다. 이 실험 결과는 제안된 방법이 다양한 심장 구조에서 효과적으로 구분되어 수행되었음을 알 수 있다. 본 연구에서 제안된 알고리즘이 의사와 방사선 의사가 영상을 판독하거나 임상 결정을 내리는데 보조적 역할을 할 것을 기대한다.

• Korean Journal of Childcare and Education
• /
• v.2 no.1
• /
• pp.89-109
• /
• 2006

The purpose of this study is to investigate the level of knowledge that mothers have on child development. To do so, a Knowledge of Child Development Inventory(KCDI) developed by Larsen &Juhasz(1986) was carried on to 156 mothers with an infant range from birth to age 24 months. The findings of the study were twofold: First, 70% of the questions on average were answered correctly by the participants. In other words, mothers generally demonstrated a high level of knowledge on child development. With regard to knowledge of child development, their knowledge on physical development of children was relatively low in comparison with that of cognitive and emotional development. In addition, mothers with a girl were more aware of child development especially in relation to child's emotional development. Also, mothers who delivered a heavier baby at birth possessed a higher level of knowledge concerning child's physical development. Second, other external factors, such as the level of family income and mother's education, played a significant role in affecting the level of knowledge on child development. Unsurprisingly, mothers with a higher income and education level demonstrated a higher level of knowledge on child's cognitive development. The findings of the study implicates that mothers seldom demonstrate a proper level of knowledge on child development. Hence, this study suggests that a practical, functional training services and resources for current and future mothers must be ensued.