• Journal of Computing Science and Engineering
• /
• 제10권4호
• /
• pp.118-127
• /
• 2016

An improved sample balanced genetic algorithm and Extreme Learning Machine (iSBGA-ELM) was designed for accurate diagnosis of Alzheimer disease (AD) and identification of biomarkers associated with AD in this paper. The proposed AD diagnosis approach uses a set of magnetic resonance imaging scans in Open Access Series of Imaging Studies (OASIS) public database to build an efficient AD classifier. The approach contains two steps: "voxels selection" based on an iSBGA and "AD classification" based on the ELM. In the first step, the proposed iSBGA searches for a robust subset of voxels with promising properties for further AD diagnosis. The robust subset of voxels chosen by iSBGA is then used to build an AD classifier based on the ELM. A robust subset of voxels keeps a high generalization performance of AD classification in various scenarios and highlights the importance of the chosen voxels for AD research. The AD classifier with maximum classification accuracy is created using an optimal subset of robust voxels. It represents the final AD diagnosis approach. Experiments with the proposed iSBGA-ELM using OASIS data set showed an average testing accuracy of 87%. Experiments clearly indicated the proposed iSBGA-ELM was efficient for AD diagnosis. It showed improvements over existing techniques.

• 한국전기전자재료학회논문지
• /
• 제23권10호
• /
• pp.820-825
• /
• 2010

In this study, we intended to design the optimal Fermi filter to apply the k-space date that is knee image of the rheumatoid arthritis patient acquired from the MRI (magnetic resonance imaging) instrument. After deciding the suitable coefficient for the Fermi filter, the results were compared with modified Fermi filter and inverse Chebyshev filter, Chebyshev filter, Elliptic filter and Butterworth filter. Firstly, in comparison to the results, the radiologist confirmed that modified Fermi filter was best decision for boundary of the rheumatoid arthritis images. The number of the black voxels of the histogram showed the quantity of the results. At the proposed filter images, numbers of the blacks voxels were statistically decreased. That meant voxels only appeared the black color were changed to others voxels color. Because the number of the total voxels was fixed, the area appeared block color could be effected to the other areas. If the modified Fermi filter were used for rheumatoid arthritis patient, the result will be better than other filters.

• 한국정보디스플레이학회:학술대회논문집
• /
• 한국정보디스플레이학회 2008년도 International Meeting on Information Display
• /
• pp.1138-1141
• /
• 2008

By defining a uniform reference point array corresponding to the 3D voxel array and abandoning voxels whose deviations from their respective reference points exceed a given tolerance, the distribution of voxels within the volumetric 3D image space gets uniform, effects of non-uniform distribution upon the image reconstructing are eased.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2921-2923
• /
• 2000

Adaptive template filtering has been proposed recently for an enhancement of signal-to-noise ratio. In some magnetic resonance images whose gray levels have relatively small dynamic ranges, e.g., T1 imaging, however, artificial stair-like artifact is observed in edge regions. This is partially due to edge enhancement effect in such voxels that contain multiple compounds at the boundaries of tissues. The gray levels of these voxels tend to change those of near voxels that contain single compound by the adaptive filtering, which exaggerate edge discontinuities. In this paper, we propose a technique to eliminate such artifact by identifying those voxels and assigning a larger template for them. Filtered images with the proposed technique show substantial visual enhancement at the edges without degradation of peak signal-to-noise ratio compared to the original adaptive template filtering for both magnetic resonance images and phantom images

• 한국CDE학회논문집
• /
• 제22권1호
• /
• pp.59-69
• /
• 2017

Internal capacity of a refrigerator is an important indicator for design and purchasing criteria. The components facing the internal space may have holes or gaps between parts. In traditional way, design engineers manually remodeled the parts to fill the holes and the gaps for enclosed boundary of the internal space. Then they calculated internal volume by subtracting the assembly of parts from its enclosing volume. However, filling holes and gaps is not an automated process requiring a plenty of labor and time. In this research, we have developed a voxel-based method to estimate the internal volume of a refrigerator automatically. It starts transforming all components facing the interior space into voxels and fills all holes and gaps automatically by the developed hole-filling algorithm to form a completely closed boundary of the assembly. Then, it identifies the boundary voxels that are facing to the internal voxels with any part of the component. After getting the intersection points between the boundary voxels and the surfaces of components, it generates the boundary surface of triangular facets with the intersection points. Finally, it estimates the internal volume by adding volume of each tetrahedron composed of a triangle of boundary surface and an arbitrary point.

• 한국정밀공학회지
• /
• 제13권6호
• /
• pp.90-101
• /
• 1996

Determination of assembly sequence of components is a key issue in assembly operation. Although a number of articles dealing with assembly sequence determination have appeared, an efficient and general methodology for complex products has yet to appear. The objective of this paper is to present the problems and models used to generate assembly sequence from design data. An essential idea of this research is to acquire a finite number of voxels from any complex geometric entity, such as 3D planar polygons, hollow spheres, cylinders. cones, tori, etc. In order to find a feasible assembly sequence, the following four steps are needed: (1) The components composing of an assembly product are identified and then the geometric entities of each component are extracted. (2) The geometric entities extracted in the first step are translated into a number of voxels. (3) All the mating or coupling relations between components are found by considering relations between voxels. (4) The components to be disassembled are determined using CCGs (Component Coupling Graph).

• 정보처리학회논문지B
• /
• 제18B권6호
• /
• pp.385-388
• /
• 2011

디지털 형태로 저장된 의료정보가 네트워크를 통하여 제약 없이 전송될 수 있게 되면서, 환자의 개인정보 관리는 의료 업계에서 중요한 주제로 부각되었다. 현재 두뇌 영상의 의료정보를 보호하는 방법은 환자의 신원을 은닉시키기 위하여 얼굴을 절삭하는 것이다. 그러나 절삭 과정에서 간혹 중요한 두뇌 조직부가 함께 절단되어 탈면 두뇌 영상은 의료 용도로 활용될 수 없게 손상을 입게 된다. 실린더 모양의 마스크를 덧붙임으로써 두뇌 영상의 중요한 모든 정보를 유지하면서 환자의 신원 정보를 은닉시키는 직접적인 방법을 제안하였다. 제안하는 두뇌 영상의 무손실 변경 방법은 중요한 영상정보가 손상되지 않음을 확인하였다. 또한 마스크로 입혀진 두뇌영상의 신원을 확인할 수 없는 사실도 증명되었다.

• 한국정밀공학회지
• /
• 제24권9호
• /
• pp.148-157
• /
• 2007

A new method which uses distance fields scheme and marching cube algorithm is proposed in order to get an accurate offset model of arbitrary shapes composed of triangular net. In the method, the space bounding the triangular net is divided into smaller cells. For the efficient calculation of distance fields, valid cells which will generate a portion of offset model are selected previously by the suggested detection algorithm. These valid cells are divided again into much smaller voxels which assure required accuracy. At each voxel distance fields are created by calculating the minimum distances between corner points of voxels and triangular net. After generating the whole distance fields, the offset surface were constructed by using the conventional marching cube algorithm together with mesh smoothing scheme. The effectiveness and validity of this new offset method was demonstrated by performing numerical experiments for the various types of triangular net.

• 한국정밀공학회지
• /
• 제24권1호
• /
• pp.64-70
• /
• 2007

A long-exposing technique (LET) has been conducted to create nanoscale patterns applicable to diverse micro-devices using two-photon polymerization (TPP). By the weakly-polymerized region via the LET, double-layered embossing patterns can be fabricated simply in a single step. The LET makes possible a voxel and its surrounding to be fully grown into more than 500 nm in lateral size and weakly-polymerized region (WPR), respectively. In the WPR. interconnecting ribs between voxels are generated, and they lead to the creation of double-layered dot patterns. Moreover, by controlling the distance between voxels, various shapes of interconnecting rib can be fabricated when the LET is applied. Various embossing patterns were fabricated to evaluate the usefulness of the proposed technique as a novel nanopatterning technique in TPP.

• 한국멀티미디어학회논문지
• /
• 제18권2호
• /
• pp.113-119
• /
• 2015

This paper suggests an efficient method for making the high resolution volexlized model from a polygonal surface object. A distinctive strength of the method is that a surface model, however complex one, can be transformed and formed an absolute voxelized solid model in a various resolution. It caused by producing a voxel by integrating the informations for the candidated voxels separately detected in each 3D-axial direction. This method reduces memory complexity by storing the information of voxels that is produced during the 2-phase volxelization(surface and inner voxelization) of a surface object in a binary file. For the computational efficiency, a parallel process using multi-threads is applied in the process of the inner voxelization, it also takes advantage of time complexity.