• 한국인터넷방송통신학회논문지
• /
• 제24권2호
• /
• pp.119-125
• /
• 2024

본 논문에서는 인공지능 주의환기 보상전략 시스템이 발달장애인의 데이터 라벨링 작업 주의집중력에 미치는 효과를 분석하였다. 주의집중력의 척도로는 세션별 작업 정확도와 작업수행량을 사용하였다. 연구 결과, 중재가 적용된 후 연구대상자 모두 자율작업 대비 주의집중력에서 유의미한 향상이 관찰되었다. 이러한 결과는 인공지능 기술이 발달장애인의 데이터 라벨링 작업 중 주의집중력 향상에 긍정적으로 작용할 수 있음을 의미한다. 본 연구는 인공지능 기술의 적용이 발달장애인의 데이터 라벨링 작업 정확도를 향상하여 학습데이터의 품질을 높일 수 있음을 보여주고 있으며, 발달장애인의 데이터라벨링 관련 직업훈련 프로그램에 중요한 시사점을 제공하리라 본다.

• Journal of information and communication convergence engineering
• /
• 제10권3호
• /
• pp.300-306
• /
• 2012

An image attention model and its application to image quality assessment are discussed in this paper. The attention model is based on rarity quantification, which is related to self-information to attract the attention in an image. It is relatively simpler than the others but results in taking more consideration of global contrasts between a pixel and the whole image. The visual attention model is used to develop a local distortion predictor, named color visual differences predictor (CVDP), in color images in order to effectively detect luminance and color distortions.

• ETRI Journal
• /
• 제43권5호
• /
• pp.775-786
• /
• 2021

To translate in real time, a simultaneous translation system should determine when to stop reading source tokens and generate target tokens corresponding to a partial source sentence read up to that point. However, conventional attention-based neural machine translation (NMT) models cannot produce translations with adequate latency in online scenarios because they wait until a source sentence is completed to compute alignment between the source and target tokens. To address this issue, we propose a reinforced learning (RL)-based attention mechanism, the reinforced attention mechanism, which allows a neural translation model to jointly train the stopping criterion and a partial translation model. The proposed attention mechanism comprises two modules, one to ensure translation quality and the other to address latency. Different from previous RL-based simultaneous translation systems, which learn the stopping criterion from a fixed NMT model, the modules can be trained jointly with a novel reward function. In our experiments, the proposed model has better translation quality and comparable latency compared to previous models.

• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.21-28
• /
• 2019

Recent years with the development of artificial intelligence and the success of the deep model, they have been deployed in all fields of computer vision. Action recognition, as an important branch of human perception and computer vision system research, has attracted more and more attention. Action recognition is a challenging task due to the special complexity of human movement, the same movement may exist between multiple individuals. The human action exists as a continuous image frame in the video, so action recognition requires more computational power than processing static images. And the simple use of the CNN network cannot achieve the desired results. Recently, the attention model has achieved good results in computer vision and natural language processing. In particular, for video action classification, after adding the attention model, it is more effective to focus on motion features and improve performance. It intuitively explains which part the model attends to when making a particular decision, which is very helpful in real applications. In this paper, we proposed a 3D dense convolutional network based on attention mechanism(ADD-Net), recognition of human motion behavior in the video.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.485-487
• /
• 2023

컴퓨터 비전의 하위 태스크(Task)인 의미론적 분할(Semantic Segmentation)은 자율주행, 해상에서 선박찾기 등 다양한 분야에서 연구되고 있다. 기존 FCN(Fully Conovlutional Networks) 기반 의미론적 분할 모델은 다운샘플링(Dowsnsampling)과정에서 공간정보의 손실이 발생하여 정확도가 하락했다. 본 논문에서는 공간정보 손실을 완화하고자 PSA(Polarized Self-attention)의 공간정보 강조 모듈을 HRNet(High-resolution Networks)의 합성곱 블록 사이에 추가한다. 실험결과 파라미터는 3.1M, GFLOPs는 3.2G 증가했으나 mIoU는 0.26% 증가했다. 공간정보가 의미론적 분할 정확도에 영향이 미치는 것을 확인했다.

• 복식
• /
• 제22권
• /
• pp.23-44
• /
• 1994

The purpose of this study was to clarify the cultural characteristic of dress and adornments by examining articles on dress adornments and related items in Korean newspaper over periods historically and objectively by means of content analysis. This study attempted to at-tain a macro-cultural view by analysing to at-tain a macro-cultural view by analysing closely the cultural characteristics of dress and adornments as a micro-cultural system through culturally based model suggested by hamilton. The two-hundreds and eighty articles on dress and adornments were selected from newspapers(most by form Chosun Ilbo and partly from Maeil Shinbo) pulished between 1920 and 1990. The results were as follows: 1. The culture of dress and adornment received much attention during the 1930's and 1960's and little during 1950's. 2. Various cultural characteristics of dress and adornments appeared on and after 1960's: reporting more foreign news items showing foreign-oriented and future-oriented features showing cultural relativism. In the 1920's and 1970's the contents of news items on dress and adornments show the most common- mass- oriented character. Foreign-oriented cultural tendencies in cloth-ing were increasing during from 1960's to 1970's but the tendencies were turned to rather tradition-oriented features on and after 1980's compare with 1960's-1970's. Advisory critical articles on dress and adornments were small in number and insignificant but compare with other periods these received much atten-tion during the 1920's and 1980's. 3. Ideological components received much at-tention on and after 1920's to 1990. Techo-nological components received much attention during 1920's and little during 1960's. The social structural components received a little attention on and after 1920's-1940's and 1990. 4. News items on women's dress and adornments received much attention from the 1920's to 1960's and news items on both men's and women's dress and adorments were in-creasing and received much attention on and after the 1970's. 5. The pragmatic cultures were mostly re-lated to techonological components and evaluative-normative culture were mostly re-lated to ideological and social structural components. In the light of these results dress and adorments as a cultural sub-system comprise a dynamic inteacting system that articulated directly with the macro-cultural system.

• 한국도서관정보학회지
• /
• 제53권2호
• /
• pp.95-115
• /
• 2022

이 연구의 목적은 국가과학기술표준분류체계의 소분류 용어를 기계학습 알고리즘을 적용하여 기술키워드 변환하는 것이 목적이다. 이를 위해 본 연구에서는 주제어 추천에 적합한 학습 알고리즘으로 AttentionMeSH를 활용했다. 원천데이터는 한국과학기술기획평가원이 정제한 2017년부터 2020년까지 4개년 연구현황 파일을 사용하였다. 학습은 과제명, 연구목표, 연구내용, 기대효과와 같이 연구내용을 잘 표현하고 있는 4개 속성을 사용했다. 그 결과 임계치(threshold)가 0.5일 때 MiF 0.6377이라는 결과가 도출됨을 확인하였다. 향후 실제 업무에 기계학습을 활용하고, 기술키워드 확보를 위해서는 용어관리체계 구축과 다양한 속성들의 데이터 확보가 필요할 것으로 보인다.

• 대한상한금궤의학회지
• /
• 제15권1호
• /
• pp.85-115
• /
• 2023

Objectives : This study reports on the cases of four ADHD patients who treated with herbal medication based on Disease Pattern Identification Diagnostic System (DPIDS). Methods : Two children and two adults were diagnosed with ADHD according to DSM-V, and Galgeunhwangryeonhwanggeum-tang was administered. The parents of the two children completed the Korean ADHD Rating Scale (K-ARS) before and after treatment, while the other patients used the Korean ADHD Self Rating Scale (K-ASRS). The patients' somatic symptoms and other issues were evaluated through interviews. Results : In the first case, the K-ARS scale exhibited a change from 28 to 3. The patient demonstrated improvements in attention, concentration, and social behavior, along with the elimination of self-injurious behavior. In the second case, the K-ARS scale showed a change from 41 to 9. The patient in the third case demonstrated improved chronic fatigue, increased attention, work efficiency, and enhanced social skills, leading to maintaining employment for over six months for the first time in their life after the treatments. Additionally, the patient in the fourth case demonstrated improved work concentration and relief in nocturia. Conclusions : In this study, four patients exhibited improvements in symptoms of attention deficit, hyperactivity, and impulsivity before and after herbal treatment, leading to enhanced academic and social relationships. As the patients progressed in their health restoration, symptoms related to the genitourinary system also showed improvement.

• 해양환경안전학회지
• /
• 제29권6호
• /
• pp.543-551
• /
• 2023

기후변화 영향으로 이상고수온, 태풍, 홍수, 가뭄 등 재난 및 안전 관리기술은 지속적으로 고도화를 요구받고 있으며, 특히 해수면 온도는 한반도 주변에서 발생되는 여름철 적조 발생과 동해안 냉수대 출현, 소멸 등에 영향을 신속하게 분석할 수 있는 중요한 인자이다. 따라서, 본 연구에서는 해수면 온도 자료를 해양 이상현상 및 연구에 적극 활용되기 위해 통계적 방법과 딥러닝 알고리즘을 적용하여 예측성능을 평가하였다. 예측에 사용된 해수면 수온자료는 흑산도 조위관측소의 2018년부터 2022년까지 자료이며, 기존 통계적 ARIMA 방법과 Long Short-Term Memory(LSTM), Gated Recurrent Unit(GRU)을 사용하였고, LSTM의 성능을 더욱 향상할 수 있는 Sequence-to-Sequence(s2s) 구조에 Attention 기법을 추가한 Attention Long Short-Term Memory (LSTM)기법을 사용하여 예측 성능 평가를 진행하였다. 평가 결과 Attention LSTM 모델이 타 모델과 비교하여 더 좋은 성능을 보였으며, Hyper parameter 튜닝을 통해 해수면 수온 성능을 개선할 수 있었다.

• 융합신호처리학회논문지
• /
• 제22권2호
• /
• pp.85-90
• /
• 2021

운전자의 안전 운전을 위한 첨단 운전자 보조시스템(ADAS; Advanced Driver Assistance System)은 자율주행 자동차에서 중요한 연구 분야 가운데 하나이다. 특히, 이전에 자동차에 부착된 영상센서를 기반으로 한 ADAS 소프트웨어는 구축 비용이 저렴하고 그 활용도가 우수하다. 본 논문에서는 선행차의 주행 상황을 인지할 수 있는 선행 차량 후미등(Tail-Lamp)의 정지등(Break-Lamp) 영역을 검출하는 알고리즘을 제안한다. 제안하는 방법은 주행 영상으로부터 객체 추적에 우수한 성능을 보이고 있는 YOLO 기술을 이용하여 자동차 객체를 추출하고, 추출된 자동차 관심 영역의 HSV 영상을 이용하여 정지등의 밝기 변화 영역을 검출한다. 그 다음 검출된 각 정지등 후보 고립영역을 라벨링하여 후보 영역들 간의 모양 대칭성을 인지하는 선택적 주의집중 모델(Selective Attention Model)을 적용하여 정지등 영역을 검출한다. 제안한 알고리즘의 성능 평가를 위하여 다양한 주행 영상에 적용하여 실험한 결과 ADAS에 적용 가능한 성공적인 검출 결과를 보였다.