• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.816-817
• /
• 2011

최근 화석 연료 사용으로 인한 대기 오염과 자원 고갈이 심각한 사회적 문제로 대두되면서 엔진 구동 시스템에서 전동기 구동 시스템으로의 차량의 패러다임 변화가 급격히 요구되고 있으며, 이런 추세는 고성능 군용 차량도 예외는 아니다. 그 안에 포함되는 핵심 부품인 견인용 전동기는 고토크, 고출력, 전 운전 범위의 고효율화 등의 운전 특성을 만족시키기 위해 IPMSM을 사용하는 것이 가장 적합하나, 마그네틱 토크뿐 아니라 릴럭턴스 토크까지 사용하며 파라미터의 비선형적 특성 등으로 인해 설계에 많은 어려움이 있다. 본 논문에서는 IPMSM의 2D-인덕턴스, 3D-철손, 4D-전압의 비선형 데이터 세트를 이용한 운전시뮬레이션을 통해 제어시 필요한 전류-토크 테이블 및 정확도 높은 효율맵을 구성하였으며, 시작품을 통해 결과의 타당성을 검증한다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.33 no.2
• /
• pp.95-101
• /
• 2015

The study planned to suggest the efficient methods of system calibration between the range and optical sensors. The simulation was performed by considering i) design of test-bed, ii) mathematical methods of system calibration and iii) locations of the sensors. The test-bed was designed by considering specifications of the range and optical sensors. Also, the error levels of each sensor were considered in the process of simulation with dataset, which was generated under these predetermined conditions. The system calibration was carried out by using the simulated dataset in two different approaches, which are single photo resection and bundle adjustment. The results from the simulation determined that the bundle adjustment method is more efficient than the single photo resection in the system calibration between range and optical sensors. For the better results, we have used the data, obtained in various locations. In a conclusion, the most efficient case was in sequence of i) the bundle adjustment with ii) the simulated dataset, which were obtained between 2m to 4m away from the test-bed.

• Journal of Korean Neurosurgical Society
• /
• v.63 no.3
• /
• pp.386-396
• /
• 2020

Objective : To generate synthetic spine magnetic resonance (MR) images from spine computed tomography (CT) using generative adversarial networks (GANs), as well as to determine the similarities between synthesized and real MR images. Methods : GANs were trained to transform spine CT image slices into spine magnetic resonance T2 weighted (MRT2) axial image slices by combining adversarial loss and voxel-wise loss. Experiments were performed using 280 pairs of lumbar spine CT scans and MRT2 images. The MRT2 images were then synthesized from 15 other spine CT scans. To evaluate whether the synthetic MR images were realistic, two radiologists, two spine surgeons, and two residents blindly classified the real and synthetic MRT2 images. Two experienced radiologists then evaluated the similarities between subdivisions of the real and synthetic MRT2 images. Quantitative analysis of the synthetic MRT2 images was performed using the mean absolute error (MAE) and peak signal-to-noise ratio (PSNR). Results : The mean overall similarity of the synthetic MRT2 images evaluated by radiologists was 80.2%. In the blind classification of the real MRT2 images, the failure rate ranged from 0% to 40%. The MAE value of each image ranged from 13.75 to 34.24 pixels (mean, 21.19 pixels), and the PSNR of each image ranged from 61.96 to 68.16 dB (mean, 64.92 dB). Conclusion : This was the first study to apply GANs to synthesize spine MR images from CT images. Despite the small dataset of 280 pairs, the synthetic MR images were relatively well implemented. Synthesis of medical images using GANs is a new paradigm of artificial intelligence application in medical imaging. We expect that synthesis of MR images from spine CT images using GANs will improve the diagnostic usefulness of CT. To better inform the clinical applications of this technique, further studies are needed involving a large dataset, a variety of pathologies, and other MR sequence of the lumbar spine.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.8
• /
• pp.23-32
• /
• 2024

NVS (Novel View Synthesis) is a field in computer vision that reconstructs new views of a scene from a set of input views. Real-time rendering and high performance are essential for NVS technology to be effectively utilized in various applications. Recently, 3D-GS (3D Gaussian Splatting) has gained popularity due to its faster training and inference times compared to those of NeRF (Neural Radiance Fields)-based methodologies. However, since 3D-GS reconstructs a 3D (Three-Dimensional) scene by splitting and cloning (Density Control) Gaussian points, the number of Gaussian points continuously increases, causing the model to become heavier as training progresses. To address this issue, we propose two methodologies: 1) Gaussian blending, an improved density control methodology that removes unnecessary Gaussian points, and 2) a performance enhancement methodology using a depth estimation model to minimize the loss in representation caused by the blending of Gaussian points. Experiments on the Tanks and Temples Dataset show that the proposed methodologies reduce the number of Gaussian points by up to 4% while maintaining performance.

• International Journal of Advanced Culture Technology
• /
• v.4 no.2
• /
• pp.1-6
• /
• 2016

Impervious surfaces are important indicators for urban development monitoring. Accurate mapping of urban impervious surfaces with observational satellites, such as VNREDSat-1, remains challenging due to the spectral diversity not captured by an individual PAN image. In this article, five multi-resolution image fusion techniques were compared for the task of classifting urban impervious surfaces. The result shows that for VNREDSat-1 dataset, UNB and Wavelet tranformation methods are the best techniques in reserving spatial and spectral information of original MS image, respectively. However, the UNB technique gives the best results when it comes to impervious surface classification, especially in the case of shadow areas included in non-impervious surface group.

• The Korean Journal of Applied Statistics
• /
• v.25 no.4
• /
• pp.661-668
• /
• 2012

For displaying multivariate numerical data on a 2D plane by the projection, principal components biplot and the GGobi are two main tools of data visualization. The biplot is very useful for capturing the global shape of the dataset, by representing $n$ observations and $p$ variables simultaneously on a single graph. The GGobi shows a dynamic movie of the images of $n$ observations projected onto a sequence of unit vectors floating on the $p$-dimensional sphere. Even though these two methods are certainly very valuable, there are drawbacks. The biplot is too condensed to describe the detailed parts of the data, and the GGobi is too burdensome for ordinary data analyses. In this paper, "the local projective display(LPD)" is proposed for visualizing multivariate numerical data. Main steps of the LDP are 1) $k$-means clustering of the data into $k$ subsets, 2) drawing $k$ principal components biplots of individual subsets, and 3) sequencing $k$ plots by Hurley's (2004) endlink algorithm for cognitive continuity.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.261-264
• /
• 2021

본 논문에서는 SR(Super-Resolution)을 계산하는데 필요한 데이터를 효율적으로 분류하고 분할하여 빠르게 SR연산을 가능하게 하는 쿼드트리 기반 최적화 기법을 제안한다. 제안하는 방법은 입력 데이터로 사용하는 연기 데이터를 다운스케일링(Downscaling)하여 쿼드트리 연산 소요 시간을 감소시키며, 이때 연기의 밀도를 이진화함으로써, 다운스케일링 과정에서 밀도가 손실되는 문제를 피한다. 학습에 사용된 데이터는 COCO 2017 Dataset이며, 인공신경망은 VGG19 기반 네트워크를 사용한다. 컨볼루션 계층을 거칠 때 데이터의 손실을 막기 위해 잔차(Residual)방식과 유사하게 이전 계층의 출력 값을 더해주며 학습한다. 결과적으로 제안하는 방법은 이전 결과 기법에 비해 약15~18배 정도의 속도향상을 얻었다.

• Analytical Science and Technology
• /
• v.28 no.4
• /
• pp.270-277
• /
• 2015

Boiling point (BP) is one of the most fundamental physicochemical properties of organic compounds to characterize and identify the thermal characteristics of target compounds. Previously developed QSPR equations, however, still had some limitation for the specific compounds, like high-energy molecules, mainly because of the lack of experimental data and less coverage. A large BP dataset of 5,923 solid organic compounds was finally secured in this study, after dedicated pre-filtration of experimental data from different sources, mostly consisting of compounds not only from common organic molecules but also from some specially used molecules, and those dataset was used to build the new BP prediction model. Various machine learning methods were performed for newly collected data based on meaningful 2D descriptor set. Results of combined check showed acceptable validity and robustness of our models, and consensus approaches of each model were also performed. Applicability domain of BP prediction model was shown based on descriptor of training set.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.2
• /
• pp.22-31
• /
• 2023

Governments around the world are enacting laws mandating explainable traceability when using AI(Artificial Intelligence) to solve real-world problems. HAI(Human-Centric Artificial Intelligence) is an approach that induces human decision-making through Human-AI collaboration. This research presents a case study that implements the Human-AI collaboration to achieve explainable traceability in governmental data analysis. The Human-AI collaboration explored in this study performs AI inferences for generating labels, followed by AI interpretation to make results more explainable and traceable. The study utilized an example dataset from the Ministry of Oceans and Fisheries to reproduce the Human-AI collaboration process used in actual policy-making, in which the Ministry of Science and ICT utilized R&D PIE(R&D Platform for Investment and Evaluation) to build a government investment portfolio.

• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.190-196
• /
• 2023

Recently, following the development of LIDAR technology which can detect distance from the object, the interest for LIDAR based 3D object detection network is getting higher. Previous networks generate inaccurate localization results due to spatial information loss during voxelization and downsampling. In this study, we propose an attention-based convergence method and a camera-LIDAR convergence system to acquire high-level features and high positional accuracy. First, by introducing the attention method into the Voxel-RCNN structure, which is a grid-based 3D object detection network, the multi-scale sparse 3D convolution feature is effectively fused to improve the performance of 3D object detection. Additionally, we propose the late-fusion mechanism for fusing outcomes in 3D object detection network and 2D object detection network to delete false positive. Comparative experiments with existing algorithms are performed using the KITTI data set, which is widely used in the field of autonomous driving. The proposed method showed performance improvement in both 2D object detection on BEV and 3D object detection. In particular, the precision was improved by about 0.54% for the car moderate class compared to Voxel-RCNN.