• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2001.04a
• /
• pp.15-19
• /
• 2001

A dedicated reverse engineering(RE) system for rapid manufacturing of human head in a 3D bust has been developed. The first step in the process is to capture the surface details of a human head and shoulder by three scanners based upon the digital moire fringe technique. Then the multiple scans captured from different angles are aligned and merged into a single polygonal mesh, and the aligned data set is refined by smoothing, subdividing or hole filling process. Finally, the refined data set is sent to a 4-axis computer numerically control(NC) machine to manufacture a replica. In this paper, we mainly describe on the algorithms and software for aligning multiple data sets. The method is based on the recently popular Iterative Closest Point(ICP) algorithm that aligns different polygonal meshes into one common coordinate system. The ICP algorithm finds the nearest positions on one scan to a collection of points on the other scan by minimizing the collective distance between different scans. We also integrate some heuristics into the ICP to enhance the aligning process. A typical example is presented to validate the system and further research work is also discussed.

• Korean Journal of Computational Design and Engineering
• /
• v.11 no.5
• /
• pp.335-343
• /
• 2006

Design changes for an original surface model are frequently required in a manufacturing area: for example, when the physical parts are modified or when the parts are partially manufactured from analogous shapes. In this case, an efficient 3D model updating method by locally adding scan data for the modified area is highly desirable. For this purpose, this paper presents a new procedure to update an initial model that is composed of combinatorial triangular facets based on a set of locally added point data. The initial surface model is first created from the initial point set by Tight Cocone, which is a water-tight surface reconstructor; and then the point cloud data for the updates is locally added onto the initial model maintaining the same coordinate system. In order to update the initial model, the special region on the initial surface that needs to be updated is recognized through the detection of the overlapping area between the initial model and the boundary of the newly added point cloud. After that, the initial surface model is eventually updated to the final output by replacing the recognized region with the newly added point cloud. The proposed method has been implemented and tested with several examples. This algorithm will be practically useful to modify the surface model with physical part changes and free-form surface design.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.4
• /
• pp.123-130
• /
• 2018

Triangulation is one of the fundamental problems in computational geometry and computer graphics community, and it has huge application areas such as 3D printing, computer-aided engineering, surface reconstruction, surface visualization, and so on. The Delaunay refinement algorithm is a well-known method to generate quality triangular meshes when point cloud and/or constrained edges are given in two- or three-dimensional space. In this paper, we propose a simple but efficient algorithm to triangulate Voronoi surfaces of Voronoi diagram of spheres in 3-dimensional Euclidean space. The proposed algorithm is based on the Ruppert's Delaunay refinement algorithm, and we modified the algorithm to be applied to the triangulation of Voronoi surfaces in two ways. First, a new method to deciding the location of a newly added vertex on the surface in 3-dimensional space is proposed. Second, a new efficient but effective way of estimating approximation error between Voronoi surface and triangulation. Because the proposed algorithm generates a triangular mesh for Voronoi surfaces with guaranteed quality, users can control the level of quality of the resulting triangulation that their application problems require. We have implemented and tested the proposed algorithm for random non-intersecting spheres, and the experimental result shows the proposed algorithm produces quality triangulations on Voronoi surfaces satisfying the quality criterion.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.142-146
• /
• 2023

We present a method for generating 3D structures and rendering objects by combining VAE (Variational Autoencoder) and GAN (Generative Adversarial Network). This approach focuses on generating and rendering 3D models with improved quality using residual learning as the learning method for the encoder. We deep stack the encoder layers to accurately reflect the features of the image and apply residual blocks to solve the problems of deep layers to improve the encoder performance. This solves the problems of gradient vanishing and exploding, which are problems when constructing a deep neural network, and creates a 3D model of improved quality. To accurately extract image features, we construct deep layers of the encoder model and apply the residual function to learning to model with more detailed information. The generated model has more detailed voxels for more accurate representation, is rendered by adding materials and lighting, and is finally converted into a mesh model. 3D models have excellent visual quality and accuracy, making them useful in various fields such as virtual reality, game development, and metaverse.

• ETRI Journal
• /
• v.46 no.2
• /
• pp.165-174
• /
• 2024

We propose a method to suppress the speckle noise and blur effects of the light field extracted from a hologram using a deep-learning technique. The light field can be extracted by bandpass filtering in the hologram's frequency domain. The extracted light field has reduced spatial resolution owing to the limited passband size of the bandpass filter and the blurring that occurs when the object is far from the hologram plane and also contains speckle noise caused by the random phase distribution of the three-dimensional object surface. These limitations degrade the reconstruction quality of the hologram resynthesized using the extracted light field. In the proposed method, a deep-learning model based on a generative adversarial network is designed to suppress speckle noise and blurring, resulting in improved quality of the light field extracted from the hologram. The model is trained using pairs of original two-dimensional images and their corresponding light-field data extracted from the complex field generated by the images. Validation of the proposed method is performed using light-field data extracted from holograms of objects with single and multiple depths and mesh-based computer-generated holograms.

• Archives of Plastic Surgery
• /
• v.38 no.6
• /
• pp.894-898
• /
• 2011

Purpose: Coverage of full-thickness large flank defect is a challenging procedure for plastic surgeons. Some authors have reported external oblique turnover muscle flap with skin grafting, inferiorly based rectus abdominis musculocutaneous flap, and two independent pedicled perforator flaps for flank reconstruction. But these flaps can cover only certain portions of the flank and may not be helpful for larger or more lateral defects. We report a case of large flank defect after resection of extraskeletal Ewing's sarcoma which is successfully reconstructed with reverse latissimus dorsi myocutaneous flap. Methods: A 24-year-old male patient had $13.0{\times}7.0{\times}14.0$ cm sized Ewing's sarcoma on his right flank area. Department of chest surgery and general surgery operation team resected the mass with 5.0 cm safety margin. Tenth, eleventh and twelfth ribs, latissimus dorsi muscle, internal and external oblique muscles and peritoneum were partially resected. The peritoneal defect was repaired with double layer of Prolene mesh by general surgeons. $24{\times}25$ cm sized soft tissue defect was noted and the authors designed reverse latissimus dorsi myocutaneous flap with $21{\times}10$ cm sized skin island on right back area. To achieve sufficient arc of rotation, the cephalic border of the origin of latissimus dorsi muscle was divided, and during this procedure, ninth intercostal vessels were also divided. The thoracodorsal vessels were ligated for 15 minutes before divided to validate sufficient vascular supply of the flap by intercostal arteries. Results: Mild congestion was found on distal portion of the skin island on the next day of operation but improved in two days with conservative management. Stitches were removed in postoperative 3 weeks. The flap was totally viable. Conclusion: The authors reconstructed large soft tissue defect on right flank area successfully with reverse latissimus dorsi myocutaneous flap even though ninth intercostal vessel that partially nourishes the flap was divided. The reverse latissimus dorsi myocutaneous flap can be used for coverage of large soft tissue defects on flank area as well as lower back area.

• Archives of Plastic Surgery
• /
• v.38 no.2
• /
• pp.135-142
• /
• 2011

Purpose: Deficiencies of the abdominal wall can be the a result of infection, surgery, trauma, or primary herniation. For abdominal wall reconstruction, synthetic materials have been shown to provide a better long-term success rate than primary fascial repair. But, synthetic materials cannot elicit angiogenesis or produce growth factor and are therefore plagued by an inability to clear infection. As a result of the inherent drawbacks of synthetic, significant effort has been spent on the identification of new bioprosthetic materials. The aim of our study is to evaluate the effectiveness of a synthetic material (PROCEED$^{(R)}$) and an ADM ($SureDerm^{TM}$) to repair abdominal wall defects in a rabbit models. Methods: We measured the tensile strength of the $SureDerm^{TM}$ and PROCEED$^{(R)}$ by a Tension meter (Instron 4482). 16 Rabbit models were assigned to this study for abdominal wall reconstruction. Abdominal defect of 8 rabbits were reconstructed by PROCEED$^{(R)}$ and the rest were reconstructed by $SureDerm^{TM}$. We assessed gross and histologic examinations for the reconstructed abdominal wall. Results: The tensile strength of $SureDerm^{TM}$ and Gore Tex$^{(R)}$ is $14.64{\pm}0.51Mpa$, $8.54{\pm}0.45Mpa$. PROCEED$^{(R)}$ was estimated above the limits of measurement. Inflammatory reaction of PROCEED$^{(R)}$ persisted for 32weeks, but $SureDerm^{TM}$ decreased after 16weeks. Vascular ingrowth into the $SureDerm^{TM}$ was seen after 32 weeks. The basement membrane of $SureDerm^{TM}$ changed into a form of pseudoperitoneum. In PROCEED$^{(R)}$, it seemed like pseudoepithelial lining was made from the fibrosis around the mesh. Conclusion: In our study, the $SureDerm^{TM}$ not only have less inflammatory reaction and presented more angiogenesis than the PROCEED$^{(R)}$, but also have pseudoperitoneum formation. It is expected that $SureDerm^{TM}$ is useful for abdominal wall reconstruction. However, a long-term study of its usage consequences are thought to be needed.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.2
• /
• pp.132-140
• /
• 2001

This paper proposes an efficient walktlu-ough animation from two images of the same scene. To make animation easily and fast, Tour Into the Picture(TIP) enables walkthrough animation from single image but lacks the reality of its foreground object when the viewpoint moves from side to side, and view morphing uses only 2D transition between two images but restricts its camera path on the line between two views. By combining advantages of these two image-based techniques, this paper suggests a new virtual navigation technique which enable natural scene transformation when the viewpoint changes in the side-to-side direction as well as in the depth direction. In our method, view morphing is employed only in foreground objects , and background scene which is perceived carelessly is mapped into cube-like 3D model as in TIP, so as to save laborious 3D reconstruction costs and improve visual realism simultaneously. To do this, we newly define a camera transformation between two images from the relationship of the spidery mesh transformation and its corresponding 3D view change. The result animation shows that our method creates a realistic 3D virtual navigation using a simple interface.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5B
• /
• pp.429-439
• /
• 2009

When Catastrophic extreme flood occurs due to dam break, the response time for flood warning is much shorter than for natural floods. Numerical models can be powerful tools to predict behaviors in flood wave propagation and to provide the information about the flooded area, wave front arrival time and water depth and so on. But flood wave propagation due to dam break can be a process of difficult mathematical characterization since the flood wave includes discontinuous flow and dry bed propagation. Nevertheless, a lot of numerical models using finite volume method have been recently developed to simulate flood inundation due to dam break. As Finite volume methods are based on the integral form of the conservation equations, finite volume model can easily capture discontinuous flows and shock wave. In this study the numerical model using Riemann approximate solvers and finite volume method applied to the conservative form for two-dimensional shallow water equation was developed. The MUSCL scheme with surface gradient method for reconstruction of conservation variables in continuity and momentum equations is used in the predictor-corrector procedure and the scheme is second order accurate both in space and time. The developed finite volume model is applied to 2D partial dam break flows and dam break flows with triangular bump and validated by comparing numerical solution with laboratory measurements data and other researcher's data.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.14 no.4
• /
• pp.525-535
• /
• 2001

In this paper, a new adaptive analysis scheme for element-free Galerkin method(EFGM) is proposed. The novel point of this scheme is that the triangular cell structure based on the Delaunay triangulation is used in the numerical integration and the node adding/removing process. In adaptive analysis with this scheme, there is no need to divide the integration cell and the memory cell structure. For the adaptive analysis of crack propagation, the reconstruction of cell structure by adding and removing the nodes on integration cells based the estimated error should be carried out at every iteration step by the Delaunay triangulation technique. This feature provides more convenient user interface that is closer to the real mesh-free nature of EFGM. The analysis error is obtained basically by calculating the difference between the values of the projected stresses and the original EFG stresses. To evaluate the performance of proposed adaptive procedure, the crack propagation behavior is investigated for several examples.