• 한국실내디자인학회논문집
• /
• 제14권5호
• /
• pp.254-261
• /
• 2005

This paper presents an attempt to develop a visual programming language-based 3D virtual reality authoring tool intended to compose virtual interior space. The rapid development of digital technology and the wide spread of the Intenet have expanded the different uses of virtual reality in a number of applications ranging from interior design to building maintenance. In particular, the construction of cyber spaces based on existing interior spaces is becoming increasingly important. Current research, however, remains at the level of converting 3D models into virtual reality models, despite practitioners' needs for structural space models. Moreover, commercial tools to build virtual reality space have the disadvantage of targeting people who have professional knowledge of computer programs and computer graphics. Accordingly, the 3D virtual reality authoring tool developed in this research - called the VESL system - enables virtual and structural space to be easily composed using intuitive and interactive visual interfaces, which are based on visual programming techniques. The VESL system also provides an XML based semantic description of interior space, to be used to describe interior space information. We anticipate that the virtual reality spaces composed by this system will be of considerable use in the fields of architecture and interior design. Further research issues identified at the end of the research include developing a converter/filter for transforming Internet virtual reality standard language, or VRML, and evaluating the application of the system for practical use.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.237-250
• /
• 2022

Structural health monitoring (SHM) plays an important role in ensuring the safety and functionality of critical civil infrastructure. In recent years, numerous researchers have conducted studies to develop computer vision and machine learning techniques for SHM purposes, offering the potential to reduce the laborious nature and improve the effectiveness of field inspections. However, high-quality vision data from various types of damaged structures is relatively difficult to obtain, because of the rare occurrence of damaged structures. The lack of data is particularly acute for fatigue crack in steel bridge girder. As a result, the lack of data for training purposes is one of the main issues that hinders wider application of these powerful techniques for SHM. To address this problem, the use of synthetic data is proposed in this article to augment real-world datasets used for training neural networks that can identify fatigue cracks in steel structures. First, random textures representing the surface of steel structures with fatigue cracks are created and mapped onto a 3D graphics model. Subsequently, this model is used to generate synthetic images for various lighting conditions and camera angles. A fully convolutional network is then trained for two cases: (1) using only real-word data, and (2) using both synthetic and real-word data. By employing synthetic data augmentation in the training process, the crack identification performance of the neural network for the test dataset is seen to improve from 35% to 40% and 49% to 62% for intersection over union (IoU) and precision, respectively, demonstrating the efficacy of the proposed approach.

• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.351-363
• /
• 2023

Visual structural inspections are an inseparable part of post-earthquake damage assessments. With unmanned aerial vehicles (UAVs) establishing a new frontier in visual inspections, there are major computational challenges in processing the collected massive amounts of high-resolution visual data. We propose twin deep learning models that can provide accurate high-resolution structural components and damage segmentation masks efficiently. The traditional approach to cope with high memory computational demands is to either uniformly downsample the raw images at the price of losing fine local details or cropping smaller parts of the images leading to a loss of global contextual information. Therefore, our twin models comprising Trainable Resizing for high-resolution Segmentation Network (TRS-Net) and DmgFormer approaches the global and local semantics from different perspectives. TRS-Net is a compound, high-resolution segmentation architecture equipped with learnable downsampler and upsampler modules to minimize information loss for optimal performance and efficiency. DmgFormer utilizes a transformer backbone and a convolutional decoder head with skip connections on a grid of crops aiming for high precision learning without downsizing. An augmented inference technique is used to boost performance further and reduce the possible loss of context due to grid cropping. Comprehensive experiments have been performed on the 3D physics-based graphics models (PBGMs) synthetic environments in the QuakeCity dataset. The proposed framework is evaluated using several metrics on three segmentation tasks: component type, component damage state, and global damage (crack, rebar, spalling). The models were developed as part of the 2^nd International Competition for Structural Health Monitoring.

• International Journal of Computer Science & Network Security
• /
• 제22권5호
• /
• pp.79-88
• /
• 2022

Modern production requires production staff who have design competence, experience and skills to work in various types of work integrated into professional activities. Possession of digital design methods significantly expands the opportunities for professional activities of qualified workers. The purpose of our study was to study the impact of pedagogical conditions on the formation of design competence of future qualified workers in a group work. We have identified a set of pedagogical conditions that promote the development of professionally oriented artistic and technical creativity of workers in the conditions of curricular and extracurricular activities, which include motivational-target, procedural-semantic, organizational-technological, and subject-oriented. It is shown that the formation of design competence is determined by motivational, informational-active and reflection criteria, which are aimed at motivational-value, cognitive, operational-active, creative, social and emotional components of this competence. The methodology of the research is highlighted, which includes the use of the following methods: determination of the personality's motivational sphere in order to identify strong and weak motives of students activity; multiple intelligence to identify students talents in the direction of practical intelligence, which is important for design competence; determining the level of creative activity to identify manifestations of students creative abilities; identifying the type of students innovative thinking in order to develop motivation for success; factor-criterion model, developed on the basis of a qualimetric approach, which is used to identify the level of design competence formation in accordance with its components. The results of the study showed that the creation of separate pedagogical conditions in the institution of vocational education and training (VET) had a positive impact on the development of design competence, which shows the potential of artistic and technical design in the development of professional creativity of future qualified workers taking into account the environmental approach.

• 지능정보연구
• /
• 제25권3호
• /
• pp.63-88
• /
• 2019

북한의 변화와 동향 파악에 대한 연구는 북한관련 정책에 대한 방향을 결정하고 북한의 행위를 예측하여 사전에 대응 할 수 있다는 측면에서 매우 중요하다. 현재까지 북한 동향에 대한 연구는 전문가를 중심으로 과거 사례를 서술적으로 분석하여, 향후에 북한의 동향을 분석하고 대응하여 왔다. 이런 전문가 서술 중심의 북한 변화 및 동향 연구에서 비정형데이터를 이용한 텍스트마이닝 분석이 더해지면 보다 과학적인 북한 동향 분석이 가능할 것이다. 특히 북한의 동향 파악과 북한의 대남 관련 행위와 연관된 연구는 통일 및 국방 분야에서 매우 유용하며 필요한 분야이다. 본 연구에서는 북한의 신문 기사 내용을 활용한 텍스트마이닝 방법으로 북한과 관련한 핵심 단어를 구축하였다. 그리고 본 연구는 김정은 집권 이후 최근의 남북관계의 극적인 관계와 변화들을 기반으로 세 개의 기간을 나누고 이 기간 내에 국내 언론에 나타난 북한과 관련성이 높은 단어들을 시계열적으로 분석한 연구이다. 북한과 관련한 주요 단어들을 세 개의 기간별로 분류하고 당시에 북한의 태도와 동향에 따라 해당 단어와 주제들의 관련성이 어떻게 변화하였는지를 파악하였다. 본 연구는 텍스트마이닝을 이용한 연구가 남북관계 및 북한의 동향을 이해하고 분석하는 방법론으로서 얼마나 유용한 것이지를 파악하는 것이었다. 앞으로 북한의 동향 분석에 대한 연구는 물론 대북관계 및 정책에 대한 방향을 결정하고, 북한의 행위를 사전에 예측하여 대응 할 수 있는 북한 리스크 측정 모델 구축을 위한 연구로 진행 될 것이다.