• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.3
• /
• pp.116-121
• /
• 2006

In this paper, we propose the dimension reduction method of multi-dimension speech feature vector for real-time adaptation procedure in various noisy environments. This method which reduces dimensions non-linearly to map the likelihood of speech feature vector and noise feature vector. The LRT(Likelihood Ratio Test) is used for classifying speech and non-speech. The results of implementation are similar to multi-dimensional speech feature vector. The results of speech recognition implementation of detected speech data are also similar to multi-dimensional(10-order dimensional MFCC(Mel-Frequency Cepstral Coefficient)) speech feature vector.

• Journal of Digital Convergence
• /
• v.18 no.5
• /
• pp.305-315
• /
• 2020

As the incidence and prevalence of dementia increases with our aging population, so does the social burden on our society, which calls for a special emphasis on need for early diagnosis. Thus, efforts are made to prevent dementia and early detection but with current diagnostic measures, these efforts appear futile. As a solution, it is crucial to integrate and standardize healthcare big data and analysis of each index. In order to increase use of large database, the Korea National Assembly passed the Data 3 Act focusing on open-access and sharing of database, but a follow-up legislation is needed a for safer utilization. In this study, we have identified number of foreign of foreign policies through review of prior researches on the topic leading to specific enforcement ordinances tailored to the Data 3 Act for safe access and utilization of database. We also aimed to establish secure process of data collection and disposal as well as governance at the national level to ensure safe utilization of healthcare big data.

• Electronics and Telecommunications Trends
• /
• v.20 no.3 s.93
• /
• pp.92-103
• /
• 2005

멀티센서 공간영상정보 통합처리 기술은 인공 위성, 항공기 등에 탑재된 여러 가지의 센서에서 취득되는 공간영상 데이터를 통합 처리하여 고품질의 공간정보 생성을 가능케 해주는 신기술이다. 기존의 개별센서 처리 기술이 가지는 개개의 장점만을 융합하여 3차원 가상현실 공간 구성, 디지털 지도 자동 제작, 방재, 환경 등의 분야에서 사용되는 공간 데이터를 생성해낼 수 있는 장점을 지닌다. 본 고에서는 멀티센서 통합처리기술의 동향과 기술의 내용에 대하여 살펴보고, 통합처리 기술이 어떻게 활용되는지에 대한 내용을 소개한다.

• Journal of Convergence for Information Technology
• /
• v.11 no.11
• /
• pp.57-65
• /
• 2021

We propose a novel method to detect abnormal data of specific symptoms using deep learning in air pollution measurement system. Existing methods generally detect abnomal data by classifying data showing unusual patterns different from the existing time series data. However, these approaches have limitations in detecting specific symptoms. In this paper, we use DeepLab V3+ model mainly used for foreground segmentation of images, whose structure has been changed to handle one-dimensional data. Instead of images, the model receives time-series data from multiple sensors and can detect data showing specific symptoms. In addition, we improve model's performance by reducing the complexity of noisy form time series data by using 'piecewise aggregation approximation'. Through the experimental results, it can be confirmed that anomaly data detection can be performed successfully.

• Conservation Science in Museum
• /
• v.22
• /
• pp.15-26
• /
• 2019

This study was focused on the three-dimensional convergence modeling that can multilaterally analyze internal and external shapes of the Sam-Chongtong Hand Canon by optical precision scanning optimized for acquiring the surface shape and X-ray CT scanning used for obtaining the internal shape. First, the scanning results were converted by compatible extension, after which three-dimensional deviation analysis was conducted to verify mutual conformities. Accordingly, most (56.98%) deviations between the two scanning models was found be ±0.1mm. This result did not influence registration and merging based on the ICP algorithm. The merged data exhibited the external surface color, detailed shapes, internal width, and structure of the hand canon. The three-dimensional model based on optical surface scanning and X-ray CT scanning can be used for traditional technique interpretation as well as digital documentation of cultural heritage. In the future, it will contribute to deliver accessible scientific information of exhibits for visitors.

• Korean Journal of Remote Sensing
• /
• v.24 no.2
• /
• pp.179-188
• /
• 2008

This study aims for generating a city spatial model required for the creation of a 3D noise map. In this study, we propose an efficient method to generate 3D models of the terrain and buildings using only a digital map and draft maps previously established without using any sensory data. The terrain model is generated by interpolating into a grid the elevation values derived from both the contour lines and the elevation point of the digital map. Building model is generated by combining the 2D building boundaries and the building elevations extracted from the digital map and the draft map, respectively. This method has been then applied to a digital map and three sets of draft maps created in the different times. covering the entire area of Yeongdeungpo-gu. The generated city spatial model has been successfully utilized for the noise analysis and the 3D visualization of the analysis results.

• Spatial Information Research
• /
• v.23 no.2
• /
• pp.49-57
• /
• 2015

Recently, Through the linkage and opening, the fusion of the spatial information, it is necessary for productive ecosystem to provide a variety of information and to increase the civil use. Depending on the economic growth, demand for quality of life and well-being has been on the increase. Spatial information service contents for the public convenience has emerged to solve the problem such as health, safety, welfare and discomfort of daily life This study aims to implement search services for a tailored residence space through the three-dimensional data modeling on cadastral information. To achieve this goal, we established the requirements for deriving a registered object by investigating recent trend with respect to existing cadastral data model and defined property and relationship. Focusing on Songpa-gu, Jamsil station in Seoul, we implemented search services for a tailored residence space for three-dimensional right analysis in conjunction with residential and commercial complex building. As a result, we derived a way to supply 3D cadastre information through open platforms (VWorld) and to represent efficiently, which is able to improve the quality of spatial information service contents for the public convenience as well as to widen utilization of information.

• The KIPS Transactions:PartB
• /
• v.9B no.5
• /
• pp.563-570
• /
• 2002

Robust extraction of 3D facial features and global motion information from 2D image sequence for the MPEG-4 SNHC face model encoding is described. The facial regions are detected from image sequence using multi-modal fusion technique that combines range, color and motion information. 23 facial features among the MPEG-4 FDP (Face Definition Parameters) are extracted automatically inside the facial region using color transform (GSCD, BWCD) and morphological processing. The extracted facial features are used to recover the 3D shape and global motion of the object using paraperspective camera model and SVD (Singular Value Decomposition) factorization method. A 3D synthetic object is designed and tested to show the performance of proposed algorithm. The recovered 3D motion information is transformed into global motion parameters of FAP (Face Animation Parameters) of the MPEG-4 to synchronize a generic face model with a real face.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.1
• /
• pp.98-112
• /
• 2017

Recently, various research has been conducted on the use of a game engine instead of a commercial geographic information system (GIS) engine for the development of 3D GIS. The advantage of the 3D game engine is that it allows developers to develop various modules according to their abilities. In particular, in the area of disasters, a wide range of alternatives for prevention as well as prediction can be presented when new analyses are attempted by combining geographic information and disaster-related information. Furthermore, 3D analysis can be an important factor in analyzing the phenomena occurring in the real 3D world because of the nature of disasters. Therefore, in this study, we tried to develop a visualization module for flood disaster information through a 3D game engine by considering the solutions for cost and manpower problems and the degree of freedom of development. Raw flood data was mapped onto spatial information and interpolation was performed for the natural display of the mapped flood information. Furthermore, we developed a module that intuitively shows dangerous areas to users by generating cumulative information in order to display multidimensional information based on this information. The results of this study are expected to enable various flood information analyses as well as quick response and countermeasures to floods.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.3
• /
• pp.136-150
• /
• 2008

Many studies have been carried out for automatic extraction of building by LiDAR data or airphoto. Combining the benefits of 3D location information data and shape information data of image can improve the accuracy. So, in this research building recognition algorithm based on contour was used to improve accuracy of building recognition by LiDAR data and elaborate building boundary recognition by airphoto. Building recognition algorithm based on contour can generate building boundary and roof structure information. Also it shows better accuracy of building detection than the existing recognition methods based on TIN or NDSM. Out of creating buffers in regular size on the building boundary which is presumed by contour, this research limits the boundary area of airphoto and elaborate building boundary to fit into edge of airphoto by double active contour. From the result of this research, 3D building boundary will be able to be detected by optimal matching on the constant range of extracted boundary in the future.