• Journal of Digital Contents Society
• /
• v.19 no.9
• /
• pp.1811-1817
• /
• 2018

The purpose of this study is to elucidate the effects of a new clinical training simulator for dental radiography using augmented reality (AR) on user learning context. To accomplish this purpose, we divided 217 dental hygiene students into two groups. The experimental group was presented with the new clinical training simulator for dental radiography using AR, and the control group was presented with task information using a textbook. The results showed that the experimental group presented the new clinical training simulator for dental radiography using AR had a higher level of self-efficacy, interest in learning, flow, and practice satisfaction compared with the control group shown the task information using a textbook. Therefore, the AR-based radiography simulator can be utilized in dental radiology practice education as an effective educational device.

• The Journal of the Korea Contents Association
• /
• v.10 no.6
• /
• pp.205-212
• /
• 2010

Interest in the augmented reality is growing and increase of Smartphone is changing people's life style. The purpose of this study is to explore application method of image restoration based on augmented reality to museum education. Museum is the proper place that audience, including students could learn culture and history of past time. This study suggests that using smartphone application which is used by image restoration with augmented reality is efficient to museum audience's understanding and interest. A game design for domestic museums is planned. Smartphone application which is used by image restoration with augmented reality also could be utilized for exploring historic sites or enjoying local festivals.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.5
• /
• pp.313-322
• /
• 2020

A large volume of data will improve the robustness of deep learning models and avoid overfitting problems. In automatic tongue segmentation, the availability of annotated tongue images is often limited because of the difficulty of collecting and labeling the tongue image datasets in reality. Data augmentation can expand the training dataset and increase the diversity of training data by using label-preserving transformations without collecting new data. In this paper, augmented tongue image datasets were developed using seven augmentation techniques such as image cropping, rotation, flipping, color transformations. Performance of the data augmentation techniques were studied using state-of-the-art transfer learning models, for instance, InceptionV3, EfficientNet, ResNet, DenseNet and etc. Our results show that geometric transformations can lead to more performance gains than color transformations and the segmentation accuracy can be increased by 5% to 20% compared with no augmentation. Furthermore, a random linear combination of geometric and color transformations augmentation dataset gives the superior segmentation performance than all other datasets and results in a better accuracy of 94.98% with InceptionV3 models.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.85-92
• /
• 2023

Deep learning applications are increasingly being leveraged for disease detection tasks in medical imaging modalities such as X-ray, Computed Tomography (CT), and Magnetic Resonance Imaging (MRI). Most data-centric deep learning challenges necessitate the use of supervised learning methodologies to attain high accuracy and to facilitate performance evaluation through comparison with the ground truth. Supervised learning mandates a substantial amount of image and label sets, however, procuring an adequate volume of medical imaging data for training is a formidable task. Various data augmentation strategies can mitigate the underfitting issue inherent in supervised learning-based models that are trained on limited medical image and label sets. This research investigates the enhancement of a deep learning-based rib fracture segmentation model and the efficacy of data augmentation techniques such as left-right flipping, rotation, and scaling. Augmented dataset with L/R flipping and rotations(30°, 60°) increased model performance, however, dataset with rotation(90°) and ⨯0.5 rescaling decreased model performance. This indicates the usage of appropriate data augmentation methods depending on datasets and tasks.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.91-98
• /
• 2023

Human-object interaction(HOI) detection requires both object detection and interaction recognition, and requires a large amount of data to learn a detection model. Current opened dataset is insufficient in scale for training model enough. In this paper, we propose an easy and effective data augmentation method called Simple Quattro Augmentation(SQA) and Random Quattro Augmentation(RQA) for human-object interaction detection. We show that our proposed method can be easily integrated into State-of-the-Art HOI detection models with HICO-DET dataset.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.93-99
• /
• 2021

The purpose of this study is to analyze recently developed tools and relevant literature in order to discuss development scheme of augmented reality-based programming tools targeting elementary school students. Literature review draws conclusion that touch mode in the mobile augmented reality is effective, especially in the environment where manipulates commands and it is required to design contents taking class environment and teaching-learning strategy into account. Such research findings indicate that augmented reality-based programming tools targeting elementary school students should be designed to increase their interest in programming in a way that when physical teaching materials or specific space are recognized, the programmed problems will be augmented to allow students to combine the commands in the augmented environment.

• Journal of The Korean Association For Science Education
• /
• v.42 no.5
• /
• pp.557-566
• /
• 2022

This study analyzed paired middle school students' verbal and physical interactions in small group learning using augmented reality. Twelve 8th graders were paired to take classes of solubility and melting/boiling points based on augmented reality. These classes were videotaped and recorded. After the classes, all the students participated in a semi-structured interview. The results were analyzed in three sections; individual statement units of verbal interaction, interaction units of verbal interaction and physical interaction. In the individual statement units of verbal interaction, the proportion of information question/explanation was found to be high. In the interaction units of verbal interaction, the proportion of simple interaction was the highest, followed by elaborated interaction. Beneath the elaborate interaction, the proportion of cumulative interaction was found to be the highest, followed by reformative interaction. In the physical interaction, writing a worksheet and gazing at a virtual object were higher. On the basis of the results, effective ways to form a proper environment in small group learning using augmented reality are discussed.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.461-466
• /
• 2021

본 연구에서는 DECO(Dictionnaire Electronique du COreen) 한국어 전자사전과 LGG(Local-Grammar Graph)에 기반한 반자동 언어데이터 증강(Semi-automatic Symbolic Propagation: SSP) 방식에 입각하여, 핀테크 분야의 CS(Customer Service) 챗봇 NLU(Natural Language Understanding)을 위한 주석 학습 데이터를 효과적으로 생성하고, 이를 기반으로 RASA 오픈 소스에서 제공하는 DIET(Dual Intent and Entity Transformer) 아키텍처를 활용하여 핀테크 CS 챗봇 NLU 시스템을 구현하였다. 실 데이터을 통해 확인된 핀테크 분야의 32가지의 토픽 유형 및 38가지의 핵심 이벤트와 10가지 담화소 구성에 따라, DECO-LGG 데이터 생성 모듈은 질의 및 불만 화행에 대한 양질의 주석 학습 데이터를 효과적으로 생성하며, 이를 의도 분류 및 Slot-filling을 위한 개체명 인식을 종합적으로 처리하는 End to End 방식의 멀티태스크 트랜스포머 모델 DIET로 학습함으로써 DIET-only F1-score 0.931(Intent)/0.865(Slot/Entity), DIET+KoBERT F1-score 0.951(Intent)/0.901(Slot/Entity)의 성능을 확인하였으며, DECO-LGG 기반의 SSP 생성 데이터의 학습 데이터로서의 효과성과 함께 KoBERT에 기반한 DIET 모델 성능의 우수성을 입증하였다.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.131-136
• /
• 2007

본 논문에서는 모바일 증강현실에서 스마트 오브젝트 인식 및 트래킹을 위한 임베디드 마커 시스템을 제안한다. 기존의 증강 현실 연구에서 주로 사용하는 마커는 임의의 패턴을 포함하고 대상 오브젝트와는 분리되어 있다. 이는 부자연스러운 시각적 장애 요인으로 작용한다. 또한 특정한 마커를 사용하기 위해 학습 과정을 거친 후 그 결과를 인식 모듈에서 일일이 등록해야 하는 번거로움이 있다. 이러한 문제점을 해결하기 위해 제안하는 임베디드 마커는 디스플레이 장치의 유무에 따라 고정형 또는 가변 형으로 분류된 스마트 오브젝트의 특성을 고려하여 오브젝트와 마커를 결합한다. 또한 통합된 학습과 인식 모듈을 통해 오브젝트의 추가 및 시스템 확장을 용이하게 한다. 제안된 시스템은 스마트 홈 테스트베드인 ubiHome 에 적용되었다. 또한 사용 성 평가를 통해 그 효용성을 분석하였다. 이러한 임베디드 마커를 사용하면 사용자는 보다 직관적으로 마커의 용도를 예측할 수 있고 대상물과의 시선을 일치시켜 자연스러운 증강현실을 경험할 수 있을 것으로 기대된다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.43-44
• /
• 2018

본 논문에서는 로봇 매니퓰레이터의 순방향 기구학 학습에 있어서 로봇의 비용적인 측면으로 인하여 실재 로봇 매니퓰레이터를 교보재로 사용하여 실습을 하기에 제한되는 환경과 교재만으로 학습하는 제한된 환경에서 학생들이 이해하는 것 뿐만 아니라 검증이나 실험을 하기가 어려운 점을 개선하기 위해서 증강현실 기반의 시뮬레이터를 제안한다 로봇 기구학에서는 주로 교재를 사용하는데 실재로 존재하는 모델보다 회전 관절(revolute joint)과 병진 관절(prismatic joint) 모형의 조합으로 모델링한다 관절의 모형을 일종의 증강현실의 마커로 사용하여 교재에서 제안하는 모델에 더해서 개인이 조합한 모델 또한 실습이 가능하도록 하는 증강현실 시뮬레이터를 제안한다.