• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.369-372
• /
• 2023

본 논문에서는 삼각형 구조로 구성된 3차원 메쉬(Mesh)에서 합성곱 신경망(Convolution Neural Network, CNN)을 응용하여 정확도가 높은 새로운 학습 표현 기법을 제시한다. 우리는 메쉬를 구성하고 있는 폴리곤의 edge와 face의 로컬 특징을 기반으로 학습을 진행한다. 일반적으로 딥러닝은 인공신경망을 수많은 계층 형태로 연결한 기법을 말하며, 주요 처리 대상은 1, 2차원 데이터 형태인 오디오 파일과 이미지였다. 인공지능에 대한 연구가 지속되면서 3차원 딥러닝이 도입되었지만, 기존의 학습과는 달리 3차원 딥러닝은 데이터의 확보가 쉽지 않다. 혼합현실과 메타버스 시장의 확대로 인해 3차원 모델링 시장이 증가하고, 기술의 발전으로 데이터를 획득할 수 있는 방법이 생겼지만, 3차원 데이터를 직접적으로 학습에 이용하는 방식으로 적용하는 것은 쉽지 않다. 그렇게 때문에 본 논문에서는 산업 현장에서 이용되는 데이터인 메쉬 구조를 폴리곤의 최소 단위인 삼각형 형태로 구성하여 학습 데이터를 구성해 기존의 방법보다 정확도가 높은 학습 기법을 제안한다.

• The KIPS Transactions:PartD
• /
• v.11D no.1
• /
• pp.115-122
• /
• 2004

Software system modeling has a goal for finding and solving system's problems by describing and analyzing system model in formal notations. Petri nets, as graphical formalism, have been used in describing and analyzing the software systems such as parallel systems, real-time system, and protocols. In the analysis of Petri nets, general system properties such as deadlock and liveness are analyzed by the reachability analysis. On the other side, specific properties such as functional requirements and constraints are checked by model-checking. However, since these analysis methods are based on enumeration of ail possible states, there nay be state explosion problem, which means that system states exponentially increase as the size of system is larger. In this paper, we propose a new method for mechanically checking system properties with avoiding state explosion problem. At first, system properties are described in property nets then the system model and the property net are composed and analyzed. In the compositional analysis, system parts irrelevant to the specific property are reduced to minimize the analysis domain of the system. And it is possible to mechanically check whether a specific property is satisfied or not.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.1
• /
• pp.244-250
• /
• 2015

Survivability is avoidance and tolerance ability of combat systems for accomplishing mission in battle field. Therefore, the combat system has to protect or minimize any damage from threats. For this reason, many modeling and simulation based studies which analyze vulnerability of the combat system by threats, are in progress to improve survivability of the combat system. In this paper, we developed a 3D penetration analysis program for survivability analysis of combat system. To do this, we applied the penetration analysis equation to threat and protection performance of tank. Also we implemented simple tank models based on 3D CAD, and tested the developed program using the implemented tank models. As a result, we verified the developed program that is possible to analyze penetration by threat and protection performance of tank and to visualize its result, based on scenarios.

• Journal of Intelligence and Information Systems
• /
• v.21 no.4
• /
• pp.79-92
• /
• 2015

Knowledge map is widely used to represent knowledge in many domains. This paper presents a method of integrating the national R&D data and assists of users to navigate the integrated data via using a knowledge map service. The knowledge map service is built by using a lightweight ontology and a topic modeling method. The national R&D data is integrated with the research project as its center, i.e., the other R&D data such as research papers, patents, and reports are connected with the research project as its outputs. The lightweight ontology is used to represent the simple relationships between the integrated data such as project-outputs relationships, document-author relationships, and document-topic relationships. Knowledge map enables us to infer further relationships such as co-author and co-topic relationships. To extract the relationships between the integrated data, a Relational Data-to-Triples transformer is implemented. Also, a topic modeling approach is introduced to extract the document-topic relationships. A triple store is used to manage and process the ontology data while preserving the network characteristics of knowledge map service. Knowledge map can be divided into two types: one is a knowledge map used in the area of knowledge management to store, manage and process the organizations' data as knowledge, the other is a knowledge map for analyzing and representing knowledge extracted from the science & technology documents. This research focuses on the latter one. In this research, a knowledge map service is introduced for integrating the national R&D data obtained from National Digital Science Library (NDSL) and National Science & Technology Information Service (NTIS), which are two major repository and service of national R&D data servicing in Korea. A lightweight ontology is used to design and build a knowledge map. Using the lightweight ontology enables us to represent and process knowledge as a simple network and it fits in with the knowledge navigation and visualization characteristics of the knowledge map. The lightweight ontology is used to represent the entities and their relationships in the knowledge maps, and an ontology repository is created to store and process the ontology. In the ontologies, researchers are implicitly connected by the national R&D data as the author relationships and the performer relationships. A knowledge map for displaying researchers' network is created, and the researchers' network is created by the co-authoring relationships of the national R&D documents and the co-participation relationships of the national R&D projects. To sum up, a knowledge map-service system based on topic modeling and ontology is introduced for processing knowledge about the national R&D data such as research projects, papers, patent, project reports, and Global Trends Briefing (GTB) data. The system has goals 1) to integrate the national R&D data obtained from NDSL and NTIS, 2) to provide a semantic & topic based information search on the integrated data, and 3) to provide a knowledge map services based on the semantic analysis and knowledge processing. The S&T information such as research papers, research reports, patents and GTB are daily updated from NDSL, and the R&D projects information including their participants and output information are updated from the NTIS. The S&T information and the national R&D information are obtained and integrated to the integrated database. Knowledge base is constructed by transforming the relational data into triples referencing R&D ontology. In addition, a topic modeling method is employed to extract the relationships between the S&T documents and topic keyword/s representing the documents. The topic modeling approach enables us to extract the relationships and topic keyword/s based on the semantics, not based on the simple keyword/s. Lastly, we show an experiment on the construction of the integrated knowledge base using the lightweight ontology and topic modeling, and the knowledge map services created based on the knowledge base are also introduced.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.455-456
• /
• 2006

This paper presents an automatic pose-normalized 3D face data acquisition method using 2D and 3D information. We propose an automatic pose-normalized 3D face acquisition method that accomplishes 3D face modeling and 3D face pose-normalization at once. The proposed method uses 2D information with AAM (Active Appearance Model) and 3D information with 3D normal vector. The 3D face modeling system consists of 2 cameras and 1 projector. In order to verify proposed pose-normalized 3D modeling method, we made an experiment for 2.5D face recognition. The experimental result shows that proposed method is robust against pose variation.

• Journal of Cadastre & Land InformatiX
• /
• v.52 no.1
• /
• pp.69-80
• /
• 2022

Drone photogrammetry generally acquires images vertically or obliquely from above, so when photographing for the purpose of three-dimensional modeling, image matching for the ground of a building and spatial accuracy of point cloud data are poor, resulting in poor 3D mesh completeness. Therefore, to overcome this, this study analyzed the spatial accuracy of each drone image by acquiring smartphone images from the ground, and evaluated the accuracy improvement and completeness of 3D mesh when the smartphone image is not combined with the drone image. As a result of the study, the horizontal (x,y) accuracy of drone photogrammetry was about 1/200,000, similar to that of traditional photogrammetry. In addition, it was analyzed that the accuracy according to the photographing method was more affected by the photographing angle of the object than the increase in the number of photos. In the case of the smartphone image combination, the accuracy was not significantly affected, but the completeness of the 3D mesh was able to obtain a 3D mesh of about LoD3 that satisfies the digital twin city standard. Therefore, it is judged that it can be sufficiently used to build a 3D model for digital twin city by combining drone images and smartphones or DSLR images taken on the ground.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.10a
• /
• pp.447-449
• /
• 2012

There are many essential elements in making the realistic 3D games, of which one of the most important is mapping. Mapping can add the details to 3D objects easily, which are impossible or difficult by modeling alone, and can reduce file data capacities to run the game engine. The theory of the mapping is to throw 2D image on 3D object correctly, but they do not match each other at the rate of 1:1 100 percent completely due to the difference of the fundamental property of them. To solve this problem, various kinds of mapping methods are developed and used. This paper will compare and analyze the characteristics, and merits and demerits of these various kinds of mapping methods and will inquire which method is the most efficient one.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.15 no.6
• /
• pp.15-25
• /
• 2011

One dimensional combustion modeling of aluminum combustion behavior is proposed. Combustion model is assumed that region consists as follows ; preheat, reaction, post reaction region. Flame speed as a function of particle size, equivalence ratio for unitary particles and fraction ratio of micro to nano particle size for binary particles were investigated for lean burn condition at 1 atm. Results were compared with experimental data. For unitary particles, flame speed increase as particle size decreases, but opposite trend with equivalence ratio. For binary particles, flame speed increases proportionally as nano particle fraction increases. For flame structure, separated or overlapping flames are observed, depending on the fraction of nano sized particles.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.621-626
• /
• 2021

In this study, data on the study site for apartment construction was acquired, and the quantity of construction waste was calculated using a drone and mobile 3D laser scanner. The accuracy of the drone was 0.034 to 0.064m in the horizontal and vertical directions, respectively, and the mobile 3D Laser Scanner showed an accuracy of 0.018 to 0.049m in the horizontal direction, respectively. These results suggest that it is possible to construct spatial information using a drone and mobile 3D laser scanner with a value within the allowable accuracy of 1:1,000 digital terrain. The volume of construction waste calculated using the mobile 3D laser scanner data was 70,797㎥. It was possible to calculate the volume on the side of the building or some facilities that appeared as shaded areas in the drone outcomes. In addition, modeling was performed for view analysis of the apartments scheduled to be constructed and the terrain-based modeling results of the surrounding buildings. In the future, data construction and accuracy evaluation using mobile 3D laser scanners will be conducted. In addition, additional research comparing existing methods and work processes will be carried out, and the efficiency of mobile 3D laser scanners in the field of spatial information construction can be presented.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.377-380
• /
• 2006

본 논문에서는 통신해양기상위성의 지상국에 구축된 고 출력증폭기에 대한 특성 모델링 및 이에 따른 입력신호에 따른 출력 신호의 상호 간섭에 대한 결과가 정리되어 있다. 고 출력증폭기는 지상국에서 위성으로 신호를 전송할 때 필요한 장비로써 비선형 특성으로 인해 입력된 신호에 대한 다양한 하모닉 성분이 발생된다. 특히 통신해양 기상위성의 지상국의 경우에는 하나의 고 출력증폭기에 대해 두개의 서로 다른 주파수의 신호를 입력해야 하는 것을 고려해야 하기 때문에 이에 대한 영향은 사전에 반드시 분석되어야 한다. 본 논문에서는 고출력증폭기의 P1dB 특성곡선에 의해 특성 모델링을 수행한 결과 및 이러한 모델을 검증하기 위해 이미 정의된 두 신호에 대한 간섭의 양을 확인하였다. 특히 고 출력 증폭기의 전단에 속하는 상향 주파수 변환기의 특성을 고려한 신호를 사용하여 실제의 통신해양기상위성으로 데이터를 전송할 때의 결과를 얻는데 중점을 두었다. 그 결과 상호간의 간섭에 의해 발생되는 신호는 실제 보내고자 하는 신호의 대역과 매우 가깝게 있으며 그 크기는 보내고자 하는 신호의 특성에 따라 달라지는 것을 확인 할 수 있었다.