• Journal of Advanced Navigation Technology
• /
• v.18 no.3
• /
• pp.201-208
• /
• 2014

Due to diversification of ships and limited space of pier, when ships come alongside the shallow water or narrow pier, sea area with small mooring facilities and floating matter is frequently applied. Through these, coming alongside the various space is capable and cost effectiveness is enhanced. However, when ships, applying small mooring facilities and floating matter, come alongside, there can be some impulse by waves between the floating things and ships which possibly leads to mass disaster. Therefore, there should be forecasts and analysis of the movement caused by waves. On this study, we develop measuring system for movement analysis of mooring and floating matters which provides base data with movement traits by measuring 3-D exercise information and acceleration at mokpo maritime university marina facility. Also, the composition and principles of the developed system is explained.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1723-1735
• /
• 2022

Landslides are one of the most prevalent natural disasters, threating both humans and property. Also landslides can cause damage at the national level, so effective prediction and prevention are essential. Research to produce a landslide susceptibility map with high accuracy is steadily being conducted, and various models have been applied to landslide susceptibility analysis. Pixel-based machine learning models such as frequency ratio models, logistic regression models, ensembles models, and Artificial Neural Networks have been mainly applied. Recent studies have shown that the kernel-based convolutional neural network (CNN) technique is effective and that the spatial characteristics of input data have a significant effect on the accuracy of landslide susceptibility mapping. For this reason, the purpose of this study is to analyze landslide vulnerability using a pixel-based deep neural network model and a patch-based convolutional neural network model. The research area was set up in Gangwon-do, including Inje, Gangneung, and Pyeongchang, where landslides occurred frequently and damaged. Landslide-related factors include slope, curvature, stream power index (SPI), topographic wetness index (TWI), topographic position index (TPI), timber diameter, timber age, lithology, land use, soil depth, soil parent material, lineament density, fault density, normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used. Landslide-related factors were built into a spatial database through data preprocessing, and landslide susceptibility map was predicted using deep neural network (DNN) and CNN models. The model and landslide susceptibility map were verified through average precision (AP) and root mean square errors (RMSE), and as a result of the verification, the patch-based CNN model showed 3.4% improved performance compared to the pixel-based DNN model. The results of this study can be used to predict landslides and are expected to serve as a scientific basis for establishing land use policies and landslide management policies.

• Tunnel and Underground Space
• /
• v.33 no.4
• /
• pp.281-298
• /
• 2023

Continuous research efforts are being devoted to unmanned mobile platforms for lunar exploration. There is an ongoing demand for real-time information processing to accurately determine the positioning and mapping of areas of interest on the lunar surface. To apply deep learning processing and analysis techniques to practical rovers, research on software integration and optimization is imperative. In this study, a foundational investigation has been conducted on real-time analysis of virtual lunar base construction site images, aimed at automatically quantifying spatial information of key objects. This study involved transitioning from an existing region-based object recognition algorithm to a boundary box-based algorithm, thus enhancing object recognition accuracy and inference speed. To facilitate extensive data-based object matching training, the Batch Hard Triplet Mining technique was introduced, and research was conducted to optimize both training and inference processes. Furthermore, an improved software system for object recognition and identical object matching was integrated, accompanied by the development of visualization software for the automatic matching of identical objects within input images. Leveraging satellite simulative captured video data for training objects and moving object-captured video data for inference, training and inference for identical object matching were successfully executed. The outcomes of this research suggest the feasibility of implementing 3D spatial information based on continuous-capture video data of mobile platforms and utilizing it for positioning objects within regions of interest. As a result, these findings are expected to contribute to the integration of an automated on-site system for video-based construction monitoring and control of significant target objects within future lunar base construction sites.

• Science of Emotion and Sensibility
• /
• v.14 no.4
• /
• pp.593-604
• /
• 2011

This study examines potential growth and different perspectives on green-walls, which are being re-evaluated recently. The aim is to identify the viewer preferences concerning the walls by collecting a wide spectrum of information to create an interior design related comprehensive and intellectual database. Also based on this proposition, choosing a green surface of indoor wall as a specific example stimulus, identifying a relationship betweeationsviewer's preference factors. The relationship will formulate detailed and absolute qualities, which will reach potential areas that the green walls can be widely applied in. As a result of a factor analysis, the viewers classified into the 3 factors as is pleasant, gorgeous and rich about the stimulus. Preferred key factors, which are closely related to emotional image language, were; cool, tidy, comfortable and beautiful. The common factors in preferred design stimulus, in order of importance are color>elements>image/form/plants. Specific levels of design factors according to relevance are contrast>furniture> modern>central/creeper foliage plants. In the meantime, this study is leading the process of quantitative measurement of green-walls to a new design direction and it is critical to consistently experimenting to back up the theory with solid evidence.

• Journal of Korean Society for Geospatial Information Science
• /
• v.22 no.4
• /
• pp.3-11
• /
• 2014

Recently, rapid urbanization has necessitated continuous updates in digital map to provide the latest and accurate information for users. However, conventional aerial photogrammetry has some restrictions on periodic updates of small areas due to high cost, and as-built drawing also brings some problems with maintaining quality. Alternatively, this paper proposes a scheme for efficient and accurate update of digital map using point cloud data acquired by Terrestrial Laser Scanner (TLS). Initially, from the whole point cloud data, the building sides are extracted and projected onto a 2D image to trace out the 2D building footprints. In order to register the footprint extractions on the digital map, 2D Affine model is used. For Affine parameter estimation, the centroids of each footprint groups are randomly chosen and matched by means of a modified RANSAC algorithm. Based on proposed algorithm, the experimental results showed that it is possible to renew digital map using building footprint extracted from TLS data.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_3
• /
• pp.809-820
• /
• 2017

Efficient management of large-size wastes generated from disasters etc. is always in demand. Large-size wastes are closely connected to the environment, producing adverse effects on the air quality, water quality, living environment and so on. When large-size wastes are generated, we must be able to estimate the generated amount in order to transfer them to a temporary trans-shipment site, or to properly treat them. Currently, we estimate the amount of generated large-size wastes by using satellite images or unit measure for wastes; however, the accuracy of such estimations have been constantly questioned. Therefore, the present study was performed to establish three-dimensional spatial information based on UAS, to measure the amount of waste, and to evaluate the accuracy of the measurement. A measurement was made at a waste site by using UAS, and the X, Y, Z RMSE values of the three-dimensional spatial information were found to be 0.022 m, 0.023 m, and 0.14 m, all of which show relatively high accuracy. The amount of waste measured using these values was computed to be approximately $4,273,400m^3$. In addition, the amount of waste at the same site was measured by using Terrestrial LiDAR, which is used for the precise measurement of geographical features, cultural properties and the like. The resulting value was $4,274,188m^3$, which is not significantly different from the amount of waste computed by using UAS. Thus, the possibility of measuring the amount of waste using UAS was confirmed, and UAS-based measurement is believed to be useful for environmental control with respect to disaster wastes, large-size wastes, and the like.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.4
• /
• pp.189-198
• /
• 2024

Lipreading is one of the important parts of speech recognition, and several studies have been conducted to improve the performance of lipreading in lipreading systems for speech recognition. Recent studies have used method to modify the model architecture of lipreading system to improve recognition performance. Unlike previous research that improve recognition performance by modifying model architecture, we aim to improve recognition performance without any change in model architecture. In order to improve the recognition performance without modifying the model architecture, we refer to the cues used in human lipreading and set other regions such as chin and cheeks as regions of interest along with the lip region, which is the existing region of interest of lipreading systems, and compare the recognition rate of each region of interest to propose the highest performing region of interest In addition, assuming that the difference in normalization results caused by the difference in interpolation method during the process of normalizing the size of the region of interest affects the recognition performance, we interpolate the same region of interest using nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation, and compare the recognition rate of each interpolation method to propose the best performing interpolation method. Each region of interest was detected by training an object detection neural network, and dynamic time warping templates were generated by normalizing each region of interest, extracting and combining features, and mapping the dimensionality reduction of the combined features into a low-dimensional space. The recognition rate was evaluated by comparing the distance between the generated dynamic time warping templates and the data mapped to the low-dimensional space. In the comparison of regions of interest, the result of the region of interest containing only the lip region showed an average recognition rate of 97.36%, which is 3.44% higher than the average recognition rate of 93.92% in the previous study, and in the comparison of interpolation methods, the bilinear interpolation method performed 97.36%, which is 14.65% higher than the nearest neighbor interpolation method and 5.55% higher than the bicubic interpolation method. The code used in this study can be found a https://github.com/haraisi2/Lipreading-Systems.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.213-224
• /
• 2009

This paper presents a new algorithm for estimating a driver's physical condition using real-time vision system and performs experimentation for real facial image data. The system relies on a face recognition to robustly track the center points and sizes of person's two pupils, and two side edge points of the mouth. The face recognition constitutes the color statistics by YUV color space together with geometrical model of a typical face. The system can classify the rotation in all viewing directions, to detect eye/mouth occlusion, eye blinking and eye closure, and to recover the three dimensional gaze of the eyes. These are utilized to determine the carelessness and drowsiness of the driver. Finally, experimental results have demonstrated the validity and the applicability of the proposed method for the estimation of a driver's physical condition.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.124-129
• /
• 2006

음악은 소리를 즐긴다는 뜻을 담고 있다. 감상자에게 단순한 청각적 자극을 넘어 즐거움을 주기 위해선 음악적인 경험이 뒷받침되어야 한다. 가상 현실을 이용한 사용자와 시스템 간의 상호작용을 음악 경험 제공에 접목하려는 시도는, 새로운 경험을 통해 일반인들이 보다 쉽게 음악을 접하고 체험함으로써 음악을 통해 즐거움을 얻을 수 있도록 도움을 주는 데에 그 목적이 있다. 가상 오케스트라를 구현하고 지휘 동작을 재현하는 것은 이러한 가능성을 극대화하는 연구이다. 본 논문에서는 가상 오케스트라를 구현하기 위해 필수적인 중간 단계로, 사용자의 지휘 동작을 감지하여 연주의 박자(속도)를 제어하는 지휘 시뮬레이션 시스템을 제시한다. 실제의 지휘 동작을 분석하고, 동작의 변화를 인식하기 위하여 가속도 센서를 이용, 공간상에서 지휘봉의 움직임을 가속도 정보로 수집하여 이에 상응하는 박자의 제어를 구현한다. 사용자의 박자 명시에 따라 변화하는 상하 방향의 가속도를 센서를 통해 전압 신호로 입력 받고, DSP 의 A/D conversion 모듈에서 디지털 신호로 변환, 일정 수준 이상의 신호를 박자 정보로 직렬통신을 통해 컴퓨터에 전달한다. 컴퓨터에서는 Max/MSP를 이용하여 각 박자 사이의 시간 간격을 측정하고 상응하는 MIDI 음악을 재생하는 방식으로 시스템이 구현된다. 기존 연구에서 사용된 CCD 카메라에 의한 Motion Tracking 을 보완하여 동작의 크기에 따라 음량을 조절한다. 본 논문에서 제시되는 시스템은 지휘 동작에서 가장 특징적으로 나타나는 상하 방향의 급격한 가속도 변화를 직접 입력 받기 때문에 기존 시스템에 비해 지휘 동작의 인식 성공률을 높일 수 있으며, 화상 처리 및 계산에 의한 지연을 최소화할 수 있다. 또한, 장치의 규모를 소형화하여 보다 지휘봉의 형태에 가까운 인터페이스를 제공하며, 적합한 응용 콘텐츠를 접목할 경우 게임 컨트롤러로의 발전 가능성이 있다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.29 no.1
• /
• pp.55-62
• /
• 2011

Amidst the raising of climate change in relation to the earth's environment as an international issue, there is a growing interest in forest resources. In particular, Korea faces a period in which we need to control carbon release pursuant to the Convention on Climate Change and the enforcement of the Kyoto Protocol; therefore, the importance of forests is becoming greater. Recently, there has been a focus on light detection and ranging (Lidar) which is a means of acquiring in a short time various necessary pieces of information for forest management as three dimensional geospatial information. In this study, the carbon absorption of a forest was measured by using the Lidar data obtained from the Lidar. Carbon absorption release was calculated on the basis of three criteria involving the minimum height of a tree, the density of the forest, and the minimum area of the forest, which are items proposed by the Forest resources surveyor. Through this study, a method of extracting the carbon absorption of a forest area using the Lidar data quantitatively was confirmed.