• 한국실내디자인학회논문집
• /
• 제22권5호
• /
• pp.171-178
• /
• 2013

In this study, we sought to find out the effect of different emotional interior images of the community facilities in an apartment complex on a youth brain wave by analyzing an Electroencephalograph (EEG). Based on the frequency of usage, we selected learning facilities, cultural facilities, and sport facilities. For brain stimulation, the visual stimulants with three different emotional words, cheerful, gentle, and elegant, were used based on I.R.I image scale. Overall, total nine different emotional images were used. Based on our findings, we conclude that: first, in order to improve learning concentration of the youth, a learning facility for the youth needs to be designed by skillfully combining the soft and comfortable colors from the gentle image and the murky and dark colors from the elegant image. Second, when designing a cultural facility, it is preferable to consider the elegant image for a calm and comfortable space. Third, a sport facility design needs to preclude dark colors and apply light colors to create a dynamic and lively space. Furthermore, we found out that the youth has established static images of each functionally different facility through their experience and learning. Therefore, it is imperative to plan community facilities in an apartment complex in a way to connect the space function with the emotional characteristics of the youth in order to support and encourage energetic activities and learning of the community youth.

• 대한임베디드공학회논문지
• /
• 제19권1호
• /
• pp.47-55
• /
• 2024

We spend a lot of time in indoor space, and the space has a huge impact on our lives. Interior design plays a significant role to make an indoor space attractive and functional. However, it should consider a lot of complex elements such as color, pattern, and material etc. With the increasing demand for interior design, there is a growing need for technologies that analyze these design elements accurately and efficiently. To address this need, this study suggests a deep learning-based design analysis system. The proposed system consists of a semantic segmentation model that classifies spatial components and an image classification model that classifies attributes such as color, pattern, and material from the segmented components. Semantic segmentation model was trained using a dataset of 30000 personal indoor interior images collected for research, and during inference, the model separate the input image pixel into 34 categories. And experiments were conducted with various backbones in order to obtain the optimal performance of the deep learning model for the collected interior dataset. Finally, the model achieved good performance of 89.05% and 0.5768 in terms of accuracy and mean intersection over union (mIoU). In classification part convolutional neural network (CNN) model which has recorded high performance in other image recognition tasks was used. To improve the performance of the classification model we suggests an approach that how to handle data that has data imbalance and vulnerable to light intensity. Using our methods, we achieve satisfactory results in classifying interior design component attributes. In this paper, we propose indoor space design analysis system that automatically analyzes and classifies the attributes of indoor images using a deep learning-based model. This analysis system, used as a core module in the A.I interior recommendation service, can help users pursuing self-interior design to complete their designs more easily and efficiently.

• 문화기술의 융합
• /
• 제7권4호
• /
• pp.635-640
• /
• 2021

본 논문에서는 촬영된 주차장 사진으로부터 빈 주차 공간을 자동 식별할 수 있는 주차 공간 자동 식별 시스템에 대해 설명한다. 이 시스템은 딥러닝 기법에 기반한 시스템으로, 다양한 주차장 사진들을 토대로 학습을 진행하여 식별 결과의 정확도가 높으며, 기존의 주차 관리 시스템에 적용할 수 있다. 한편, 본 시스템은 손쉬운 적용 테스트를 위해, 스마트폰용 애플리케이션으로도 개발되었다. 따라서 스마트폰 카메라를 통해 주차장 사진을 찍으면, 촬영된 이미지를 자동 인식하며 빈 주차 공간을 자동 식별할 수 있다.

• 지능정보연구
• /
• 제27권3호
• /
• pp.175-197
• /
• 2021

최근 딥 러닝 기술의 발전으로 뉴스, 블로그 등 다양한 문서에 포함된 텍스트 분석에 딥 러닝 기술을 활용하는 연구가 활발하게 수행되고 있다. 다양한 텍스트 분석 응용 가운데, 텍스트 분류는 학계와 업계에서 가장 많이 활용되는 대표적인 기술이다. 텍스트 분류의 활용 예로는 정답 레이블이 하나만 존재하는 이진 클래스 분류와 다중 클래스 분류, 그리고 정답 레이블이 여러 개 존재하는 다중 레이블 분류 등이 있다. 특히, 다중 레이블 분류는 여러 개의 정답 레이블이 존재한다는 특성 때문에 일반적인 분류와는 상이한 학습 방법이 요구된다. 또한, 다중 레이블 분류 문제는 레이블과 클래스의 개수가 증가할수록 예측의 난이도가 상승한다는 측면에서 데이터 과학 분야의 난제로 여겨지고 있다. 따라서 이를 해결하기 위해 다수의 레이블을 압축한 후 압축된 레이블을 예측하고, 예측된 압축 레이블을 원래 레이블로 복원하는 레이블 임베딩이 많이 활용되고 있다. 대표적으로 딥 러닝 모델인 오토인코더 기반 레이블 임베딩이 이러한 목적으로 사용되고 있지만, 이러한 기법은 클래스의 수가 무수히 많은 고차원 레이블 공간을 저차원 잠재 레이블 공간으로 압축할 때 많은 정보 손실을 야기한다는 한계가 있다. 이에 본 연구에서는 오토인코더의 인코더와 디코더 각각에 스킵 연결을 추가하여, 고차원 레이블 공간의 압축 과정에서 정보 손실을 최소화할 수 있는 레이블 임베딩 방법을 제안한다. 또한 학술연구정보서비스인 'RISS'에서 수집한 학술논문 4,675건에 대해 각 논문의 초록으로부터 해당 논문의 다중 키워드를 예측하는 실험을 수행한 결과, 제안 방법론이 기존의 일반 오토인코더 기반 레이블 임베딩 기법에 비해 정확도, 정밀도, 재현율, 그리고 F1 점수 등 모든 측면에서 우수한 성능을 나타냄을 확인하였다.

• 한국인터넷방송통신학회논문지
• /
• 제13권6호
• /
• pp.229-235
• /
• 2013

본 논문은 각종 센서와 관리자 등을 통해 온도와 조도를 측정하여 전력량을 모니터링하며 제어하고 있는 공장용 에너지관리시스템(FEMS, Factory Energy Management System), 빌딩용 에너지관리시스템(BEMS, Building Energy Management System), 주택용 에너지관리시스템(HEMS, Home Energy Management System)등으로 크게 나누어지는 기존의 에너지관리시스템(EMS : Energy Management System)에서 사용하고 있는 각종 센서 정보들을 포함한 수집 가능한 빅 데이터를 활용하여 본 논문에서 제안하는 공간 기반 에너지관리시스템(SEMS, Space-based Energy Management System)의 추론엔진을 통해 일정한 크기와 유사한 특성을 가진 단위 공간을 정의하고 에너지를 효율적으로 관리하기 위한 공간의 크기나 특성에 따라서 SEMS의 추론엔진의 Self-Learning을 통해 추론엔진 자신이 학습을 통해 점차 스마트하게 진화하면서, 사용되는 전력량을 절감하는 방안을 제시하고자 한다.

• 대한건축학회논문집:계획계
• /
• 제34권12호
• /
• pp.13-20
• /
• 2018

This research is to analyze the public cases of information facilities in terms of central circulations in multi level volumes such as atrium or court which provide visual intervention between different spaces and physical connections such as bridges. Hunt Library design balances the understood pre-existing needs with the University's emerging needs to create a forward-thinking learning environment. While clearly a contemporary structure within a traditional context of the NCSU campus, the Hunt Library provides a positive platform for influencing its surroundings. Both technical and programmatic innovations are celebrated as part of the learning experience and provide a versatile and stimulating environment for students. Public library as open spaces connecting to an interactive social domain over communities can provide variety of learning environments, or technology based labs. There are many cases of the public information spaces with dynamic networks where participants can play their roles in physical space as well as in the intellectual stimulation. In the research, new public projects provide typologies of information spaces with user oriented media. The research is to address a creative transition between the reading space and the experimental links of the integration of state-of-the-art technology is highly visible in the building's design. The user-friendly browsing system that replaces the traditional browsing with the virtual shelves classified and archived by their form, is to reduce the storage space of the public library and it is to allow more space for collaborative learning. In addition to the intelligent robot of information storages, innovative features is the large-scale visualization space that supports team experiments to carry out collaborative online works and therefore the public library's various programs is to provide visitors with more efficient participatory environment.

• 예술인문사회 융합 멀티미디어 논문지
• /
• 제6권3호
• /
• pp.291-300
• /
• 2016

스마트 디바이스와 사물 인터넷의 등장은 온라인과 오프라인의 경계를 허무는 O2O 서비스의 등장으로 이어졌다. 이는 오프라인 시장에 온라인 서비스의 강점이 덧붙여지면서 오프라인 공간이 디지털화가 됨을 의미하며, 오프라인 산업의 판도를 바꾸고 있다. 이러한 오프라인 시장의 변화 양상과는 다르게 전시 산업은 오프라인 산업에서 꾸준한 성장세를 보이고 있으나, 전시 산업 또한 O2O 서비스와의 접목으로 새로운 부가가치를 창출이 가능한 것으로 보았다. 본 논문은 코엑스에서 열린 '2015 서울 디자인 페스티벌'에서 20명을 대상으로 설문을 진행하였다. 설문은 공간 구조에 대한 분석 용도 및 기계학습을 위한 데이터 세트를 생성하는데 사용되었다. 본 논문은 기존의 공간 구조에 대한 분석연구가 가진 문제점을 파악하여 공간 구조에 대한 새로운 분석 방법을 제안하였다. 또한 생성된 데이터 세트를 기반으로 기계학습을 진행하여 전시 공간 내 O2O 서비스 배치를 위한 평가 모델을 제안한다.

• 교육시설 논문지
• /
• 제22권3호
• /
• pp.3-11
• /
• 2015

Under the establishment and operating regulations of school presented only minimum standard for the founding of the school. Therefore, It is intended to suggest the reasonable space standard of the school facility, through comparison & analysis of facility standard in each city and provincial education office. Especially, The facilities standard of Chungcheongbuk-do Office Of Education has been exceeded standardization of architectural space which is proposed the Ministry of Education, and they has made the various learning space reflected creative ideas by designing all of the new school competition since 2000. In order to deal with reacting the changing method of studying like as the examples of Middle & High school in Chungcheongbuk-do, they need to set aside the enough required area for a head of students and the common space which is more than 40% in the total area according to the various learning space, securing the supporting facilities, and break & movement of the students. Moreover, Each of the city and provincial education offices are needed to establish the standardization of proper area for space organization of the planned school throughout upcoming competition.

• 천문학회보
• /
• 제44권1호
• /
• pp.51.2-51.2
• /
• 2019

Multi-wavelength observations become very popular in astronomy. Even though there are some correlations among different sensor images, it is not easy to translate from one to the other one. In this study, we apply a deep learning method for image-to-image translation, based on conditional generative adversarial networks (cGANs), to solar images. To examine the validity of the method for scientific data, we consider several different types of pairs: (1) Generation of SDO/EUV images from SDO/HMI magnetograms, (2) Generation of backside magnetograms from STEREO/EUVI images, (3) Generation of EUV & X-ray images from Carrington sunspot drawing, and (4) Generation of solar magnetograms from Ca II images. It is very impressive that AI-generated ones are quite consistent with actual ones. In addition, we apply the convolution neural network to the forecast of solar flares and find that our method is better than the conventional method. Our study also shows that the forecast of solar proton flux profiles using Long and Short Term Memory method is better than the autoregressive method. We will discuss several applications of these methodologies for scientific research.

• 천문학회보
• /
• 제44권2호
• /
• pp.74.2-74.2
• /
• 2019

We present a possible bias in the estimation of velocity dispersions for galaxy groups due to the contribution of subgroups which are infalling into the groups. We execute a systematic search for flux-limited galaxy groups and subgroups based on the spectroscopic galaxies with r < 17.77 mag of SDSS data release 12, by using DBSCAN (Density-Based Spatial Clustering of Application with Noise) and Hierarchical Clustering Method which are well known unsupervised machine learning algorithm. A total of 2042 groups with at least 10 members are found and ~20% of groups have subgroups. We found that the estimation of velocity dispersions of groups using total galaxies including those in subgroups are underestimated by ~10% compared to the case of using only galaxies in main groups. This result suggests that the subgroups should be properly considered for mass measurement of galaxy groups based on the velocity dispersion.