• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.586-589
• /
• 2024

본 실험 연구에서는 주의 메커니즘과 컨볼루션 신경망을 결합하여 모델을 개선하는 방법을 탐색하는 딥 러닝 기술을 소개한다. 이 기술은 지도 학습 방식을 위해 공개 데이터 세트의 쓰레기 분류 데이터를 사용하고, Grad-CAM 기술과 채널 주의 메커니즘 SE 를 적용하여 모델의 분류 의사 결정 과정을 더 잘 이해하기 위해 히트 맵을 생성한다. Grad-CAM 기술을 사용하여 히트 맵을 생성하면 분류 중에 모델이 집중하는 영역을 시각화할 수 있다. 이는 모델의 분류 결정을 설명하는 방법을 제공하여 다양한 이미지 카테고리에 대한 모델 결정의 기초를 더 잘 이해할 수 있다. 실험 결과는 전통적인 합성곱 신경망과 비교하여 제안한 방법이 쓰레기 분류 작업에서 더나은 성능을 달성한다는 것을 보여준다. 주의 메커니즘과 히트맵 해석을 결합함으로써 우리 모델은분류 정확도를 향상시킬 수 있다. 이는 실제 응용 분야의 이미지 분류 작업에 큰 의미가 있으며 해석 가능성에 대한 딥 러닝 연구 진행을 촉진하는 데 도움이 된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.329-331
• /
• 2012

본 논문에서는 영상에서의 시각적 자극의 특징에 의한 돌출과 특정 대상에 관련한 단서들간의 상호작용에 기반하여 얼굴을 검출하는 주의집중모델을 제안한다. 제안하는 모델은 얼굴에 대한 하향식 다중 단서로 모양(shape), 피부색(skin color), 밝기(luminance), 거리에 대응하는 크기, 깊이 등을 사용하며 이들 단서들이 상향식 프로세스와의 상호작용을 통해 목표하는 얼굴을 검출하도록 유도하는 상향식/하향식 결합에 기반한다. 제안하는 방법은 크기 및 회전변화를 갖는 다수의 얼굴을 포함한 영상에서 얼굴검출을 수행함으로써 성능을 검증하였다.

• Journal of KIISE
• /
• v.42 no.5
• /
• pp.609-620
• /
• 2015

Viewers of videos are likely to absorb more information from the part of the screen that attracts visual attention. This fact has led to the visual attention models that are being used in producing and evaluating videos. In this paper, we investigate the factors that are significant to visual attention and the mathematical form of the visual attention model. We then estimated the visual attention probability using the statistical design of experiments. The analysis of variance (ANOVA) verifies that the motion velocity, distance from the screen, and amount of defocus blur affect human visual attention significantly. Using the response surface modeling (RSM), we created a visual attention score model that concerns the three factors, from which we calculate the visual attention probabilities (VAPs) of image pixels. The VAPs are directly applied to existing gradient based 3D effect perception measurement. By giving weights according to our VAPs, our algorithm achieves more accurate measurement than the existing method. The performance of the proposed measurement is assessed by comparing them with subjective evaluation as well as with existing methods. The comparison verifies that the proposed measurement outperforms the existing ones.

• Journal of the Korean Society for information Management
• /
• v.39 no.2
• /
• pp.35-60
• /
• 2022

This study reports on the motivational use of special libraries from a user's perspective that investigated four motivational components: attention, relevance, confidence, and satisfaction (ARCS) of the library users. Even though special libraries received relatively little attention compared to public libraries, special libraries could fill the niche of public libraries for the special and specific needs of the community in the current library environments in South Korea. Qualitative data were collected through individual interviews and forty people participated in the project. The findings of this study show the four themes (ARCS) that users need motivation as a way to start using a special library, thereby staying focused and revisiting the library. It is possible that the ARCS model will contribute to the implementation, application, and practice of both special libraries and their services in the library environment. The results are expected to expand our knowledge on the motivational ARCS use of special libraries and to serve as basic data when designing motivational strategies and plans of the systems for special libraries.

• The Journal of the Korea Contents Association
• /
• v.18 no.6
• /
• pp.131-143
• /
• 2018

The purpose of this study is to investigate whether ego-resilience has mediating effects on the relationships between elementary school student's smartphone dependence and attention. This study was conducted on 326 5th grade elementary school students in Gimhae city, and SPSS 24.0 program was used for data analysis. The result of this study was as follows. First, elementary school students' smartphone dependence showed a significant negative correlation with ego-resilience. Second, the elementary school students' smartphone dependence showed a significant negative correlation with attention. Third, as a result of analyzing the mediating effect of ego-resilience on the elementary school students' relationship with their smartphone dependence and attention, the ego-resilience has a partial mediating effect. Based on these results, suggestions were as follows: education for good use of smartphone, establishment and dissemination of family model related to smartphone use, and practical efforts and training to improve ego-resilience.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.605-608
• /
• 2021

KoBERT는 한국어 자연어처리 분야에서 우수한 성능과 확장성으로 인해 높은 위상을 가진다. 하지만 내부에서 이뤄지는 연산과 패턴에 대해선 아직까지 많은 부분이 소명되지 않은 채 사용되고 있다. 본 연구에서는 KoBERT의 핵심 요소인 self-attention의 패턴을 4가지로 분류하며 특수 토큰에 가중치가 집중되는 현상을 조명한다. 특수 토큰의 attention score를 층별로 추출해 변화 양상을 보이고, 해당 토큰의 역할을 attention 매커니즘과 연관지어 해석한다. 이를 뒷받침하기 위해 한국어 분류 작업에서의 실험을 수행하고 정량적 분석과 함께 특수 토큰이 갖는 의미론적 가치를 평가한다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.153-156
• /
• 2007

인간의 시각 시스템은 선택적 주의 집중에 의해 시각 수용체로 도달되는 많은 물체들 중에서 필요한 정보만을 추출하여 원하는 작업을 수행한다. Itti와 Koch는 시각적 주의를 제어할 수 있는, 신경계를 모방한 계산적 모델을 제안하였으나 조명환경에 고정적인 saliency map을 구성하였다. 따라서, 본 논문에서는 영상에서 ROI(region of interest)을 탐지하기 위한 조명환경에 적응적인 saliency map 모델을 구성하는 기법을 제시한다. 변화하는 환경에서 원하는 특징을 부각시키기 위하여 상황에 적응적인 동적 가중치를 부여한다. 동적 가중치는 conspicuity map에 S.K. Chang이 제안한 PIM(Picture Information Measure)을 적용시켜 정보량을 측정한 후, 이에 따라 정규화된 값을 부여함으로써 구현한다. 제안하는 조명환경에 강인한 적응적인 saliency map 모델 구현의 성능을 얼굴검출 실험을 통하여 검증하였다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.123-132
• /
• 2017

Traffic lights contain very important information for safety driving. So, the delivery of the information to drivers in real-time is a very critical issue for advanced driver assistance systems. However, traffic light detection is quite difficult because of the small sized traffic lights and the occlusion in real world. In this paper, a traffic light detection method using modified color based saliency map and morphological information is proposed. It shows 98.14% of precisions and 83.52% of recalls on computer simulations.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.4
• /
• pp.239-247
• /
• 2013

This study is aimed to perform the qualitative evaluation based on the ARCS Model of educational mobile applications for smart phones. The evaluation has been performed targeting 60 students who attending the subject of informational education in 3 different universities in 2012 by allowing them to select the available educational mobile App installed in their smart phone. After, the level of persistent learning-intension from each student and the efficacy of ARCS motivational strategy was measured at learner's perspective. The structural equation model(SEM) was established and analyzed with PLS method to understand the relationship between the ARCS motivational strategy and the persistent learning-intension. The results of the study could be summarized as followings. First, the educational mobile App in various the motivational strategies showed different results that is the highest attention as well as the lowest satisfactory level. Second, the relevance in motivation strategies had the statistically significant effect in attention, confidence, and satisfaction. On the other hand, the other factors of attention, relevance, and confidence showed statistically significant effect in satisfaction. Finally, result demonstrate that the relevance is the critical factor inducing the significant effect in persistent learning-intension among the motivational strategies.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.2
• /
• pp.95-102
• /
• 2004

In this paper, Experience Sensitive Cumulative Neural Network (ESCNN) is introduced, which can cumulate the same or similar experiences. As the same or similar training patterns are cumulated in the network, the system recognizes more important information in the training patterns. The functions of forgetting less important information and attending more important information resided in the training patterns are surveyed and implemented by simulations. The system behaves well under the noisy circumstances due to its forgetting and/or attending properties, even in 50 percents noisy environments. This paper also describes the creation of the generalized patterns for the input training patterns.