• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.44 no.1
• /
• pp.108-114
• /
• 2007

This paper presented an image reconstruction method based on the original capsule endoscopy photos yielding a 2-D image for faster diagnosis proposes. The proposed method constructed a 3-D intestine model using the massive images obtained from the capsule endoscope. It merged all images and completed a 3-D model of an intestine. This 3-D model was reformed as a 2-D plane image showing the inner side of the entire intestine. The proposed image composition was evaluated by the 3-D simulator, OpenGL. This approach was demonstrated successfully. A physician can find the location of a disease at a glance because the composite image provided an easy-to-understand view to show the patient's intestine and thereby shorten diagnosis time.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.4
• /
• pp.294-301
• /
• 2016

In this paper, we present methods to efficiently parallelize iterative 3D image reconstruction by exploiting trigeneous devices (three different types of device) at the same time: a CPU, an integrated GPU, and a discrete GPU. We first present a technique that exploits single instruction multiple data (SIMD) architectures in GPUs. Then, we propose a performance estimation model, based on which we can easily find the optimal data partitioning on trigeneous devices. We found that the performance significantly varies by up to 6.23 times, depending on how SIMD units in GPUs are accessed. Then, by using trigeneous devices and the proposed estimation models, we achieve optimal partitioning and throughput, which corresponds to a 9.4% further improvement, compared to discrete GPU-only execution.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2009.11a
• /
• pp.99-102
• /
• 2009

본 연구에서는 실제 기증받은 시체를 이용하여 인체의 각 구조물을 실제 모습 그대로 3D 이미지화하는 과정에 대해 알아보았다. 인체의 구조물을 3D 로 이미지화하는 과정은 다음과 같다. 먼저 시체를 0.2mm 간격으로 절단하여 절단면의 사진을 찍은 후, 각 절단면의 사진에서 각각의 구조물을 구역화하여 색칠을 한 후, 구역화한 이미지에서 외곽선을 추출하여 벡터 이미지를 만든다. 이 외곽선을 1mm 간격으로 쌓아 올린 후 그 표면을 재구성하여 3D 이미지로 변환하는 과정으로 진행되었다. 3D 이미지의 제작은 가슴 부위에 한정하여 이루어졌다. 인체의 해부학적인 모형을 3D 이미지로 시각화함으로써 얻는 효과는 일반인을 대상으로 인체의 내부에 대한 시각적인 호기심을 충족시켜주고 의학 상식을 넓히는데 도움을 줄 수 있을 것 이다. 또한 의대생들을 비롯한 의학 전문가들에게는 생생한 해부학 강의용으로도 활용 가능하다. 향후 Haptic 시스템을 이용한 의료 실습 어플리케이션과 접목될 수도 있을것이고, fMRI 데이터를 비롯한 타 데이터와의 융합을 통해 시각화하여 서비스 할 수도 있다. 이처럼 인체의 3D 모형은 의료분야에서 광범위하게 활용될 수 있는 데이터로써 그 가치를 지닐 것이다.

• Journal of IKEEE
• /
• v.16 no.2
• /
• pp.95-101
• /
• 2012

A novel approach for the reconstruction of 3D building model from aerial image data is proposed in this paper. In this approach, a Centroid Neural Network (CNN) with a metric of line segments is proposed for connecting low-level linear structures. After the straight lines are extracted from an edge image using the CNN, rectangular boundaries are then found by using an edge-based grouping approach. In order to avoid producing unrealistic building models from grouping lined segments, a hierarchical grouping method is proposed in this paper. The proposed hierarchical grouping method is evaluated with a set of aerial image data in the experiment. The results show that the proposed method can be successfully applied for the reconstruction of 3D building model from satellite images.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1987.04a
• /
• pp.27-30
• /
• 1987

Some modification of laplacian pyramidal coding have been done and tested for a test image, From our studies, its apperar that 3-D redictor which exploits inter and intra plane redundancy simultaneously somewhat imporves the image quality, We also have tried to reduce the bit rate by only sending zeroth plane image values that correstpond to estimated dege points which can be obtained from the first plane For both cases the feedback of quantization errof in the previous plane has been in the reconstruction of each plane. Subjective and SNR tests show the better performance of the studied methods oner the conventional one.

• Journal of the Korean Society for Precision Engineering
• /
• v.26 no.12
• /
• pp.138-145
• /
• 2009

Custom medical treatment is being widely adapted to lots of medical applications. A technology for 3D modeling is strongly required to fabricate medical implants for individual patient. Needs on true 3D CAD data of a patient is strongly required for tissue engineering and human body simulations. Medical imaging devices show human inner section and 3D volume rendering images of human organs. CT or MRI is one of the popular imaging devices for that use. However, those image data is not sufficient to use for medical fabrication or simulation. This paper mainly deals how to generate 3D geometry data from those medical images. A new image processing technology is introduced to reconstruct 3D geometry of a human body vacancy from the medical images. Then a surface geometry data is reconstructed by using Marching cube algorithm. Resulting CAD data is a custom 3D geometry data of human vacancy. This paper introduces a novel 3D reconstruction process and shows some typical examples with implemented software.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.7
• /
• pp.326-334
• /
• 2024

Extra-oral imaging techniques such as Panoramic X-rays (PXs) and Cone Beam Computed Tomography (CBCT) are the most preferred imaging modalities in dental clinics owing to its patient convenience during imaging as well as their ability to visualize entire teeth information. PXs are preferred for routine clinical treatments and CBCTs for complex surgeries and implant treatments. However, PXs are limited by the lack of third dimensional spatial information whereas CBCTs inflict high radiation exposure to patient. When a PX is already available, it is beneficial to reconstruct the 3D oral structure from the PX to avoid further expenses and radiation dose. In this paper, we propose 3DentAI - an U-Net based deep learning framework for 3D reconstruction of oral structure from a PX image. Our framework consists of three module - a reconstruction module based on attention U-Net for estimating depth from a PX image, a realignment module for aligning the predicted flattened volume to the shape of jaw using a predefined focal trough and ray data, and lastly a refinement module based on 3D U-Net for interpolating the missing information to obtain a smooth representation of oral cavity. Synthetic PXs obtained from CBCT by ray tracing and rendering were used to train the networks without the need of paired PX and CBCT datasets. Our method, trained and tested on a diverse datasets of 600 patients, achieved superior performance to GAN-based models even with low computational complexity.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.1003-1004
• /
• 2008

In this paper, we propose a 3D modeling method using Laser Slit Beam and Stereo Camera. We can get depth information of image by analyzing projected Laser Slit Beam on object. 3D modeling is demanded exquisite merge of 3D data. In our approach, we can get the depth image where the reliability is high. Each reconstructed 3D modeling is combined by the sink information which is acquired by SIFT (Scale Invariant Feature Transform) Algorithm. We perform experiments using indoor images. The results show that the proposed method works well in indoor environments

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.391-394
• /
• 2005

In this paper, we present an approach that is able to reconstruct 3 dimensional metric models from un-calibrated images acquired by a freely moved camera system. If nothing is known of the calibration of either camera, nor the arrangement of one camera which respect to the other, then the projective reconstruction will have projective distortion which expressed by an arbitrary projective transformation. The distortion on the reconstruction is removed from projection to metric through self-calibration. The self-calibration requires no information about the camera matrices, or information about the scene geometry. Self-calibration is the process of determining internal camera parameters directly from multiply un-calibrated images. Self-calibration avoids the onerous task of calibrating cameras which needs to use special calibration objects. The root of the method is setting a uniquely fixed conic(absolute quadric) in 3D space. And it can make possible to figure out some way from the images. Once absolute quadric is identified, the metric geometry can be computed. We compared reconstruction image from calibrated images with the result by self-calibration method.

• Journal of the Korean Society of Radiology
• /
• v.5 no.5
• /
• pp.255-260
• /
• 2011

We wanted to evaluate the usefulness of three-dimensional reconstructive images using computed tomography for rib fracture patients. The reconstruction used in clinical multi planar reformation(MPR), volume rendering technique(VRT), and image data using quantitative methods and qualitative methods were compared. Much more, the artifact shadow was minimized to reconstruct with 3D volumetric image by using an law data in the analysis of the reconstructive image and chest CT scan of the evaluation result fractures of the thoracic patient. And we could know that the fractures of the thoracic determination and three dimension volume image reconstruction time were reduced.