• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.391-395
• /
• 2004

본 논문은 맞춤형 신발제작을 위하여 맞춤형 신발에 필요한 화형제작용 데이터를 위한 3차원 측정 장치를 통해 획득한 발의 형상을 인공지능 기법을 기반으로 하는 최적화된 형상을 복원하는 방법을 제시하고자 한다. 본 연구를 위해 개발된 시스템은 PC를 기반으로 하는 기존의 3차원측정 방식을 이용하여 상, 하, 좌, 우로 각각 장착된 8대의 CCD 카메라와 4대의 레이져를 통해 화형 및 발의 형상 데이터를 획득한다 획득된 데이터들은 인공지능 기법을 이용한 영상처리 알고리즘으로 처리되며, 처리 결과는 기존의 지능 기법을 도입하지 않은 시스템에 비해 노이즈제거 특성이 향상되었고, 후처리과정을 간소화 할 수 있다. 따라서 본 논문에서는 3차원 측정을 위해 기구적인 부분과 하드웨어적인 부분의 시스템을 구성하고, 데이터 처리용 소프트웨어에서 입력영상의 전처리 과정 중 영상의 이진화 단계에서 임계값을 결정하기 위하여 간단한 신경망을 사용하였으며, 이에 대한 결과를 제시하고자 한다.

• Journal of Korean Ophthalmic Optics Society
• /
• v.12 no.4
• /
• pp.87-90
• /
• 2007

To manufacture the lens mold used in producing polycarbonate (PC) lenses, the core manufacturing is needed and this core manufacturing is generally performed by diamond turning machine (DTM) or computer numerical control (CNC) lathe. The numerical data about the lens core feature is necessarily needed for operating of these devices. Therefore, we developed the program which calculate the numerical data about the lens core feature. The program was composed to be able to input aspherical coefficients of lens feature, display the graph of lens feature, and save the numerical data file.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.127-134
• /
• 2010

In continuous vocabulary recognition system by probability distribution of clustering method has used model parameters of an advance estimate to generated each contexts for phoneme data surely needed but it has it's bad points of gaussian model the accuracy unsecure of composed model for phoneme data. To improve suggested probability distribution mixed gaussian model to optimized that phoneme data search supported configuration thread system. This paper of configuration thread system has used extension facet classification user phoneme configuration thread information offered gaussian model the accuracy secure. System performance as a result of represent vocabulary dependence recognition rate of 98.31%, vocabulary independence recognition rate of 97.63%.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.586-589
• /
• 2017

소프트웨어 프로젝트 관리를 위한 여러 항목 중, 형상 관리는 매우 중요한 영역이다. 눈에 보이지 않는 소프트웨어의 결과물을 산출물인 문서로 설명, 이해, 관리하는 활동이기 때문이다. 형상 관리는 현재 한국항공우주연구원에서 운영 중인 지구관측위성에서도 중요하게 여겨지고 있다. 위성 영상 데이터의 효율적인 관리 등을 위해 형상 관리의 필요성이 계속해서 대두되어 왔다. 본 논문에서는, 기존에 다루어지는 일련의 과정들을 체계적으로 정리 및 새롭게 정의하고, 더불어 향상 관리 범위에 포함되어 있는 관리자 및 사용자가 필요로 하는 각 산출 문서의 내용을 편리하게 사용할 수 있도록 지원하는 Node.js 기반의 형상 관리 시스템 개발 방안에 대하여 기술하고자 한다. 아울러, 이 Node.js 플랫폼이 View단과 Server단을 하나로 통합하여 개발하는 방식이기 때문에, 우리가 필요로 하는 위성 영상 데이터의 효율적인 향상 관리 지원을 위한 시스템 개발에 적용해 보았다. 결과적으로, 제안하는 설계 방법이 적합하다는 것을 단계적 단위 시험을 통하여 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.316-318
• /
• 2001

본 논문에서는 자기조직화 형상지도(Self-organizing Feature Maps)를 사용하여 움직이는 물체에 대해 움직임의 특성을 자동으로 분석하였다. Kohonen Network는 자기조직을 형성하는 unsupervised learning 알고리즘으로서, 이 논문에서는 생태계에서의 데이터를 Patternizing하고, Clustering 하는데 사용한다. 본 논문에서 Kohonen 신경망의 학습에 사용한 데이터는 CCD 카메라로 물고기의 움직임을 추적한 좌표 데이터이며, diazinon 0.1 ppm을 처리한 물고기 점 데이터와 처리하지 않은 점 데이터를 각각 낮.밤 약 10시간동안 수집하여, \circled1처리전 낮 데이터 \circled2처리전 밤 데이터 \circled3처리전 낮 데이터 \circled4처리후 밤 데이터 각각 4개의 group으로 분류한 후, Kohonen Network을 사용하여 물고기의 행동 차이를 분석하였다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.1
• /
• pp.35-41
• /
• 2020

3D laser scanners are an effective way to quickly acquire a large amount of data about an object. Recently, it is used in various fields such as surveying, displacement measurement, 3D data generation of objects, construction of indoor spatial information, and BIM(Building Information Model). In order to utilize the point cloud data acquired through the 3D laser scanner, it is necessary to make the data acquired from many stations through a matching process into one data with a unified coordinate system. However, analytical researches on the accuracy of point cloud data according to the registration method are insufficient. In this study, we tried to analyze the accuracy of registration method of point cloud data acquired through 3D laser scanner. The point cloud data of the study area was acquired by 3D laser scanner, the point cloud data was registered by the ICP(Iterative Closest Point) method and the shape registration method through the data processing, and the accuracy was analyzed by comparing with the total station survey results. As a result of the accuracy evaluation, the ICP and the shape registration method showed 0.002m~0.005m and 0.002m~0.009m difference with the total station performance, respectively, and each registration method showed a deviation of less than 0.01m. Each registration method showed less than 0.01m of variation in the experimental results, which satisfies the 1: 1,000 digital accuracy and it is suggested that the registration of point cloud data using ICP and shape matching can be utilized for constructing spatial information. In the future, matching of point cloud data by shape registration method will contribute to productivity improvement by reducing target installation in the process of building spatial information using 3D laser scanner.

• The KIPS Transactions:PartA
• /
• v.11A no.7 s.91
• /
• pp.555-562
• /
• 2004

Both global volume reduction and local shape changes of hippocampus within the brain indicate their abnormal neurological states. Hippocampal shape analysis consists of two main steps. First, construct a hippocampal shape representation model ; second, compute a shape similarity from this representation. This paper proposes a novel method for the analysis of hippocampal shape using integrated Octree-based representation, containing meshes, voxels, and skeletons. First of all, we create multi-level meshes by applying the Marching Cube algorithm to the hippocampal region segmented from MR images. This model is converted to intermediate binary voxel representation. And we extract the 3D skeleton from these voxels using the slice-based skeletonization method. Then, in order to acquire multiresolutional shape representation, we store hierarchically the meshes, voxels, skeletons comprised in nodes of the Octree, and we extract the sample meshes using the ray-tracing based mesh sampling technique. Finally, as a similarity measure between the shapes, we compute $L_2$ Norm and Hausdorff distance for each sam-pled mesh pair by shooting the rays fired from the extracted skeleton. As we use a mouse picking interface for analyzing a local shape inter-actively, we provide an interaction and multiresolution based analysis for the local shape changes. In this paper, our experiment shows that our approach is robust to the rotation and the scale, especially effective to discriminate the changes between local shapes of hippocampus and more-over to increase the speed of analysis without degrading accuracy by using a hierarchical level-of-detail approach.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.1222-1225
• /
• 2013

오늘날 기업의 시스템은 훨씬 복잡해지고 다루는 데이터의 규모는 점점 대형화됨에 따라 환경 변화에 기업이 대응하기가 어려워지고 있다. IT 환경의 변화에 기업이 대응하기 위해서는 메타데이터 관리시스템의 기능을 개선할 필요가 있다. 본 논문에서는 데이터 모델링에 대한 새로운 요구사항을 신속히 반영하기 위해서 모델의 변경 이력을 참조하지 않고 이관해야 할 목표 데이터베이스의 테이블의 형상을 비교하여 이관 관리 프로세스를 향상시킨 알고리즘을 설계하였다. 본 논문서 설계한 알고리즘을 통해 모델의 변경 이력을 참조할 경우 발생하는 데이터베이스 변경의 처리 횟수를 줄임으로써 프로세스 개선의 효과를 기대할 수 있는 알고리즘을 제시하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.11
• /
• pp.494-500
• /
• 2020

Owing to the complexity of shapes and elements, some difficulties are found in the modeling and sharing phases in a project at the earlier design stages. This paper extends the boundaries by suggesting the data interoperability between 3D modeling software, McNeel Rhino 3D and BIM system, and Autodesk® Revit® Architecture. The main research methodology is to link the architectural form data in the NURBS supporting the 3DM format, especially for integrating surface properties into the mass family template of Revit. This algorithm-driven interoperability approach using visual programming, such as Dynamo in conjunction with Autodesk®, can be applicable in a theoretical part and also a practical use-case. This paper summarizes these results as sequence guidelines and project template recommendations suggesting an efficient design process to interoperate geometric data with the BIM system to manipulate and control the regular and curved form of office buildings.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.21 no.3
• /
• pp.233-238
• /
• 2008

In this paper, a shape design optimization method for linearly elastic problems is developed using isogeometric approach. In many design optimization problems for practical engineering models, initial raw data usually come from a CAD modeler. Then, designers should convert the CAD data into finite element mesh data since most of conventional design optimization tools are based on finite element analysis. During this conversion, there are some numerical errors due to geometric approximation, which causes accuracy problems in response as well as design sensitivity analyses. As a remedy for this phenomenon, the isogeometric analysis method can be one of the promising approaches for the shape design optimization. The main idea of isogeometric approach is that the basis functions used in analysis is exactly the same as the ones representing the geometry. This geometrically exact model can be used in the shape sensitivity analysis and design optimization as well. Therefore the shape design sensitivity with high accuracy can be obtained, which is very essential for a gradient-based optimization. Through numerical examples, it is verified that the shape design optimization based on an isogeometic approach works well.