• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.257-260
• /
• 2020

본 논문은 단일의 옷감 이미지로 가상의 그래픽 렌더링을 위해 Pix2Pix 방법을 이용하여 Normal map 을 생성하는 방법을 제시한다. 구체적으로 단일의 이미지를 이용해서 Normal map 를 생성하기 위해, Color image 와 Normal map 쌍의 training dataset 을 Pix2Pix 방법을 이용해서 학습시킨다 또한, test dataset 의 Color image 를 입력으로 넣어 생성된 Normal map 결과를 확인한다. 그리고 선행연구에서 사용되어오던 U-Net 방식의 방법과 본 논문에서 사용한 Pix2Pix 를 이용한 Normal map 생성 결과를 SSIM(Structural Similarity Index)으로 비교 평가한다. 또한, 생성된 Normal map 을 렌더링하고자 하는 가상 객체의 사이즈에 맞게 사이즈를 조정하여 OpenGL 로 렌더링한 결과를 확인한다. 본 논문을 통해서 단일의 패턴 이미지를 Pix2Pix 로 생성한 Normal map 으로 옷감의 디테일을 사실감 있게 표현할 수 있음을 확인할 수 있었다.

• Journal of Broadcast Engineering
• /
• v.13 no.5
• /
• pp.596-601
• /
• 2008

In this paper, we propose a new scheme to generate region-of-interest (ROI) enhanced depth maps in the hybrid camera system, which is composed of a low-resolution depth camera and a high-resolution stereoscopic camera. The proposed method creates an ROI depth map for the left image by carrying out a three-dimensional (3-D) warping operation onto the depth information obtained from the depth camera. Then, we generate a background depth map for the left image by applying a stereo matching algorithm onto the left and right images captured by the stereoscopic camera. Finally, we merge the ROI map with the background one to create the final depth map. The proposed method provides higher quality depth information on ROI than the previous methods.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.2
• /
• pp.45-53
• /
• 2019

RGB-D SLAM (Simultaneous Localization and Mapping) refers to the technology of using deep camera as a visual sensor for SLAM. In view of the disadvantages of high cost and indefinite scale in the construction of maps for laser sensors and traditional single and binocular cameras, a method for creating three-dimensional map of indoor environment with deep environment data combined with RGB-D SLAM scheme is studied. The method uses a mobile robot system equipped with a consumer-grade RGB-D sensor (Kinect) to acquire depth data, and then creates indoor three-dimensional point cloud maps in real time through key technologies such as positioning point generation, closed-loop detection, and map construction. The actual field experiment results show that the average error of the point cloud map created by the algorithm is 0.0045m, which ensures the stability of the construction using deep data and can accurately create real-time three-dimensional maps of indoor unknown environment.

• Journal of Industrial Convergence
• /
• v.16 no.1
• /
• pp.35-40
• /
• 2018

The procedural generation algorithm is an algorithm that automatically generates a content to be used in a game by repeatedly executing a series of rules. As the size of the game increases, the amount of content used in the game increases. Accordingly, artificial intelligence research is actively conducted to automatically generate game contents using game artificial intelligence such as procedural generation. In this paper, we propose an algorithm to create 3D game world using procedural generation. The proposed algorithm generates a two-dimensional contour in which the path is naturally connected using Perlin-Noise whose noise is gradually changed. A three-dimensional Height-Map is created based on the generated two-dimensional contour lines. The generated Height-Map show that the shape of the map is normal and that the player is able to move around all the sections as in the game world created by hand. In the future, we will improve the performance of algorithms and apply them to game.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.478-485
• /
• 2021

An injection locking technology of a semiconductor laser is a promising technology to generate optical complex signals by adjusting optical injection parameters. The extraction of the precise injection parameters plays a key role in the generation of the optical complex signal. Rate equations of semiconductor lasers under optical injection are commonly used to map the injection parameters and the corresponding optical complex signal. The accuracy of the generated optical complex signal on the injection parameters is limited since the rate equations require a locking map-based interpolation method. We propose a novel analytic method, namely rate equation-based direct extraction method, to directly calculate the injection parameters without relying on the locking map-based interpolation method. We achieved 103-times improvement of the signal accuracy by using the proposed method compared to locking-map based interpolation method.

• Bulletin of the Korean Chemical Society
• /
• v.13 no.3
• /
• pp.268-274
• /
• 1992

An attempt to improve the second harmonic generation (SHG) efficiency of MAP (methyl (2,4-dinitrophenyl)aminopropanoate) by modifying the substituents on the amino group of MAP is described. Several MAP analogues have been prepared using optically active amino acids alanine, phenylalanine and serine, and their SHG efficiencies measured. None of the MAP analogues exhibited SHG efficiencies as high as that of MAP. X-ray crystal structures of three MAP analogues have been determined. In the crystal structures of two of them, which were the derivatives of phenylalanine, two crystallographically-independent molecules existing in the asymmetric unit are aligned almost antiparallel. These structures are consistent with the very low SHG efficiencies of these compounds. On the other hand, the crystal structure of a serine derivative reveals substantial alignment of the dinitroaniline chromophore along the polar axis. However, the angle of 86.2° between the molecular charge tranfer axis and the polar axis of the crystal is still far away from the optimum value of 54.74° for the phase-matchable SHG. The structure is consistent with the SHG efficiency of this compound which is much higher than those of the phenylalanine derivatives but still lower than that of MAP. This study demonstrates the importance of the orientation of molecules in the crystal lattice in determining secod-order nonlinear optical properties of crystalline materials.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.04a
• /
• pp.263-266
• /
• 2004

We make DEM of the area that elevation value varies rapidly by the aid. This study evaluates the accuracy and workability between the existing DEM making mathod by processing break line and the DEM absorption method by using the program like ARC TIN or AML. The object data of DEM generation is 1/5,000 digital map publicated by NGII and this study uses 100pieces of map as the criteria. We correct the error by Geoconv and generate DEM by using ARC TIN, ARC VIEW. Accuracy Evaluation accomplished by drawing 100 points from 1/5000 digital map.

• Transactions of Materials Processing
• /
• v.13 no.4
• /
• pp.334-341
• /
• 2004

In this study, a remeshing algorithm adapted to the mesh density map using the Delaunay mesh generation method is developed. In the finite element simulation of forging process, the numerical error increases as the process goes on because of discrete property of the finite elements and distortion of elements. Especially, in the region where stresses and strains are concentrated, the numerical error will be highly increased. However, it is not desirable to use a uniformly fine mesh in the whole domain. Therefore, it is necessary to reduce the analysis error by constructing locally refined mesh at the region where the error is concentrated such as at the die corner. In this paper, the point insertion algorithm is used and the mesh size is controlled by using a mesh density map constructed with a posteriori error estimation. An optimized smoothing technique is adopted to have smooth distribution of the mesh and improve the mesh element quality.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.12 no.5
• /
• pp.549-556
• /
• 2009

One of the basic technology for implementing the autonomy of UGV(Unmanned Ground Vehicle) is a path planning algorithm using obstacle and raw terrain information which are gathered from perception sensors such as stereo camera and laser scanner. In this paper, we propose a generation method of DVGM(Directional Velocity Grid Map) which have traverse speed of UGV for the five heading directions except the rear one. The fuzzy system is designed to generate a resonable traveling speed for DVGM from current patch to the next one by using terrain slope, roughness and obstacle information extracted from raw world model data. A simulation is conducted with world model data sampled from real terrain so as to verify the performance of proposed fuzzy inference system.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.495-500
• /
• 2002

Recently, a service robot industry outstands as an up and coming industry of the next generation. Specially, there are so many research in self-steering movement(SSM). In order to implement SSM, robot must effectively recognize all around, detect objects and make a surrounding map with sensors. So, many robots have a sonar and a infrared sensor, etc. But, in these sensors, We only know informations about between the robot and the object as well as resolution faculty is of inferior quality. In this paper, we will introduce new algorithm that recognizes objects around robot and makes a two dimension surrounding map with a omni-direction vision camera and two stereo vision cameras.