• Korean Journal of Geomatics
• /
• 제3권2호
• /
• pp.123-128
• /
• 2004

As the applied fields of GIS are expanded to the transportation, developing internet-based applications for transportation information is getting attention increasingly. Most applications developed so far are primarily focused on guidance systems for owner-driven cars. Although some recent ones are devoted to public transportation systems, they show limitations in dealing with the following aspects: (i) people may change transportation means not only within the same type but also among different modes such as between buses and subways, and (ii) the system should take into account the time taken in transfer from one mode to the other. This study suggest the framework for developing a public transportation guidance system that generates optimized paths in the transportation network of mixed means including buses, subways and other modes. For this study, the Genetic Algorithms are used to find the best routes that take into account transfer time and other service-time constraints.

• KIEAE Journal
• /
• 제3권2호
• /
• pp.51-58
• /
• 2003

The purpose of this study is to present rational methods of multi-criteria optimization of the shape of energy saving buildings. The object is to determine the optimum dimension of the shape of a building, based on the following criteria: minimum building costs (including the cost of materials and construction) and yearly heating costs. Mathematical model described heat losses and gains in a building during the heating season. It takes into consideration heat losses through wall, roof, floor and windows. Particular attention was paid to have a more detailed description of heat gains due to solar radiation. On the assumption that shape of building is rectangle in order to solve the problem, the proportions of wall length and building height are determined by using non-linear programing methods(Kuhn-Tucker Conditions). The results constitute information for designers on the optimum proportions of wall lengths, height, and the ratios of window to wall areas for energy saving buildings.

• 한국광학회:학술대회논문집
• /
• 한국광학회 2001년도 제12회 정기총회 및 01년도 동계학술발표회
• /
• pp.34-35
• /
• 2001

The wavelength demultiplexer is an essential component in optical transmission systems using wavelength-division multiplexing(WDM), which can increase the number of channels and information capacity of optical fibers. For optical telecommunication, much attention has been given to demultiplexing two wavelengths in the 1.3${\mu}{\textrm}{m}$ of low dispersion band and 1.55${\mu}{\textrm}{m}$ of low loss window. Various integrated-optical devices have been proposed to perform this function, including conventional directional couplers, asymmetric Y-branching devices, asymmetric Mach-Zehnder interferometers and two-mode interference devices. (omitted)

• 한국실내디자인학회논문집
• /
• 제15권3호
• /
• pp.111-117
• /
• 2006

In the recent years, domestic construction companies have tried to differentiate their apartments using brand positioning. Respective construction companies are attracting public attention with discriminative brands. This study investigated residents' preferences focusing on apartments facade design used in marketing strategy of brand apartments. Four concepts in surveyed brand apartments are deluxe, modern, eco-friendly and comfortable. According to the related study on design factor of apartment facade, the design factors are chosen such as the overall building shape, balcony form, main color, exterior wall shape, material, main access shape, pent house design and window size. As a result of analysis, there was a difference between brand concepts and consumers' views. Therefore, apartment facade design to differentiate their own companies' brand images should consider the perception of customers.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2004년도 봄 학술발표논문집 Vol.31 No.1 (B)
• /
• pp.808-810
• /
• 2004

본 논문에서는 Contrast map과 Salient point를 이용하여 영상에서 중요한 객체를 자동으로 추출하는 방법을 제안한다. 우선 인간의 시각 체계와 유사한 밝기(luminance), 색상(color) 그리고 방향성(orientation) 3가지의 특징정보를 이용하여 각각의 특징정보로부터 feature map을 생성하고 이 3가지의 feature map을 선형 결합하여 contrast map을 생성한다. 이렇게 생성된 하나의 contrast map을 이용하여 대략적인 Attention Window (AW)의 위치를 결정한다. 다음으로, 영상으로부터 웨이블릿 변환을 적용하여 salient point를 찾고, salient point의 분포와 contrast map의 중요도에 따라 AW의 크기를 실제 중요 객체의 크기와 가장 유사하도록 축소시킨다. 이렇게 선택되고 축소된 AW안에서 실제 중요 객체를 추출하기 위해 AW 내부에 존재하는 영상에 대해서만 영상 분할을 하고 불필요한 영역을 제거하여 자동으로 중요객체를 추출하도록 한다.

• 스마트미디어저널
• /
• 제7권4호
• /
• pp.79-89
• /
• 2018

In recent years, falls among elderly people have gained serious attention as a major cause of injuries. Falls often lead to fatal consequences due to lack of prompt response and rescue. Therefore, a more accurate fall detection system and an effective feature extraction technique are required to prevent and reduce the risk of such incidents. In this paper, we proposed an efficient feature extraction technique based on multiple sliding windows and validated it through a series of experiments using supervised learning algorithms. The experiments were conducted using the public datasets obtained from tri-axial accelerometers. The results depicted that extraction of the feature from adjacent sliding windows led to high accuracy in supervised machine learning-based fall detection. Also, the experiments conducted in this study suggested that the best accuracy can be achieved by keeping the window size as small as 2 seconds. With the kNN classifier and dataset from wearable sensors, the experiments achieved accuracy rates of 94%.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• 한국정보통신학회논문지
• /
• 제13권6호
• /
• pp.1194-1199
• /
• 2009

스테레오 비전 시스템(stereo vision system)은 2차원 영상정보를 이용하여 3차원 깊이 정보를 획득하는 데 유용한 방법으로, 그동안 많은 연구가 진행되었다. 3차원 깊이 정보를 획득하기 위해서는 영상의 대응점을 찾아야 하는데, 속도와 정확성을 동시에 만족시키기가 어렵다. 이러한 문제를 해결하기 위해, 본 논문에서는 적응적 가중치(weight)를 적용한 cost aggregation 기반의 스테레오 정합 기법을 제안한다. 이 방법은 스테레오 영상의 특징을 이용하여 가중치를 획득하고, 색상정보, 밝기정보에 거리정보를 이용하여 가중치를 적용한 후, 이를 이용하여 대응점을 찾아 깊이 정보를 추출한다. 제안된 방법의 성능을 평가하기 위하여 ground truth가 존재하는 다양한 스테레오 영상을 이용하여 실험하였으며, 실험 결과 다양한 영상에서도 향상된 결과를 보였다.

• 한국항공우주학회지
• /
• 제48권8호
• /
• pp.631-640
• /
• 2020

현재 항공기 시뮬레이터는 실제 항공기를 이용한 조종사 훈련에 비해 경제적, 시간적 및 공간적 비용 측면에서 상당한 이점을 가지고 있어서 많은 관심을 받고 있다. 항공기 시뮬레이터의 구성 요소 중 비행역학 모델은 실제 항공기의 비행을 모의하는데 중요한 역할을 한다. 따라서 자동화된 도구를 사용하여 비행역학 모델의 충실도를 검증하는 것은 중요한 의미를 지닌다. 본 논문에서는 항공기 시뮬레이터의 효율적인 개발을 위하여 비행역학 모델의 충실도를 자동 검증하는 소프트웨어를 개발한다. 충실도 검증 프로세스로부터 도출된 요구사항에 대하여 소프트웨어 구조 및 GUI를 설계한 다음 Window 기반 환경에서 C# 언어를 이용하여 소프트웨어를 구현한다. CTSW 기종에 대한 실험 결과는 개발된 소프트웨어가 기능 및 성능, 그리고 사용자 편의성 측면에서 효과적임을 보여준다.

• 센서학회지
• /
• 제22권2호
• /
• pp.150-155
• /
• 2013

Recently, various hearing aids are developed to overcome hearing loss. There are available hearing aids, such as air conduction hearing aid, implantable middle ear hearing aid and so on. But air conduction hearing aid is inconvenience caused by howling, and ossicle chain driving type implantable middle ear hearing aid has some week point due to problem of possible nercobiosis of coupling spot along incus long process. In recent years, in order to improve these shortcomings round window (RW) driving hearing aid has been paying attention. In this paper, the physical cochlear model is proposed for a performance evaluation of the RW driving hearing aids of a transducer. In order to verify an experiment proposed on a performance of physical cochlear model, the transducer which has ossicles characteristics is used. By measuring and comparing the frequency characteristics of transducer with ossicles and human temporal bone, performance of physical cochlear model was verified. As from the result of experiment, it is expected that an implemented cochlear model is useful for evaluating characteristics of RW transducer.