• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.673-680
• /
• 2003

In this paper, we study the enhancement of VQ (Vector Quantization) design for text independent speaker recognition. In a concrete way, we present a non-iterative method which makes a vector quantization codebook and this method performs non-iterative learning so that the computational complexity is epochally reduced The proposed Classified Space VQ (CSVQ) design method for text Independent speaker recognition is generalized from Semi-noniterative VQ design method for text dependent speaker recognition. CSVQ contrasts with the existing desiEn method which uses the iterative learninE algorithm for every traininE speaker. The characteristics of a CSVQ design is as follows. First, the proposed method performs the non-iterative learning by using a Classified Space Codebook. Second, a quantization region of each speaker is equivalent for the quantization region of a Classified Space Codebook. And the quantization point of each speaker is the optimal point for the statistical distribution of each speaker in a quantization region of a Classified Space Codebook. Third, Classified Space Codebook (CSC) is constructed through Sample Vector Formation Method (CSVQ1, 2) and Hyper-Lattice Formation Method (CSVQ 3). In the numerical experiment, we use the 12th met-cepstrum feature vectors of 10 speakers and compare it with the existing method, changing the codebook size from 16 to 128 for each Classified Space Codebook. The recognition rate of the proposed method is 100% for CSVQ1, 2. It is equal to the recognition rate of the existing method. Therefore the proposed CSVQ design method is, reducing computational complexity and maintaining the recognition rate, new alternative proposal and CSVQ with CSC can be applied to a general purpose recognition.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1996.10a
• /
• pp.61-65
• /
• 1996

본 논문은, 공간영역에서 화상을 압축할 수 있는 Quad-tree 부호화법을 분석하고, 보다 화질 및 압축율을 개선하기 위하여 적응 불록분할 및 병합 알고리듭을 제안하엿다. 화상은 에지부분을 제외하고는 인접한 화소들간에 데이터의 용장도가 높으므로 이 영역을 하나의 대표값으로 설정하여 그 값과 그 블록의 위치좌표를 부호화할 수 있다. Quad-tree 분할은 초기의 병합을 제외하고 순차적으로 분할과정만 반복처리하지만 본 알고리듬에서는 단위블록(3$\times$3 호소) 의 평균잘류에너지(MRE)를 이용하여 블록의 분할과 병합을 반복처리한다. 시뮬레이션결과, 본 알고리듭은 압축율 1bit/pixel에서 기존의 Quad-tree 방법보다 PSNR에서 1.0dB의 개선이 있었으며, 화상의 블록화 현상도 전혀 나타나지 않았다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.39 no.3
• /
• pp.253-260
• /
• 2015

Particle contamination in a cleanroom is very complex with a complicated process and several pieces of spreading equipment. Detailed information on the locations of the contamination sources and the path of the contamination is needed for economical and efficient control of the contaminant particles in such a cleanroom. An allocation method was developed to quantitatively predict the contamination generated from the pollution sources. In this paper, we propose a zoning method to accelerate the computation time for estimating the contributions. Our results showed that we can quantitatively estimate the amount of contamination generated from pollution sources.

• The KIPS Transactions:PartB
• /
• v.15B no.5
• /
• pp.407-412
• /
• 2008

Color segmentation is one of the most challenging problems in image processing especially in case of handling the images with cluttered background. Great amount of color segmentation methods have been developed and applied to real problems. In this paper, we suggest a new methodology. Our approach is focused on background extraction, as a complimentary operation to standard foreground object segmentation, using self-organizing feature selective property of unsupervised self-learning paradigm based on the competitive algorithm. The results of our studies show that background segmentation can be achievable in efficient manner.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.22 no.2
• /
• pp.153-161
• /
• 2009

A domain/boundary decomposition method is applied for efficient analyses of thermo-elasto-viscoplastic damage and contact problems under the assumption of infinitesimal deformation. For the decomposition of a whole domain and contact boundaries, all the equality constraints on the interface and contact interfaces are restated with simple penalty functional. Therefore, the non-linearity of the problem is localized within finite element matrices in a few subdomains and on contact interfaces. By setting up suitable solution algorithms, the computational efficiency can be improved considerably. The general tendency of the computational efficiency is illustrated with some numerical experiments.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1995.10a
• /
• pp.19-24
• /
• 1995

단순한 형상의 소음기는 평면파이론에 의해 비교적 간단하게 음향성능을 해석적으로 구할 수 있다. 그러나 소음기의 형상이 복잡해지거나 해석하고자 하는 주파수의 범위가 평면파의 차단주파수 이상이 될 경우 소음기 내부의 음장이 평면파에서 벗어나게 되어 평면파 이론에 의한 해석은 실제와 상당한 오차가 발생하게 되므로 음장에 대한 3차원 해석이 필요하다. 이론적으로 3차원 문제를 해석할 수 있는 경우는 형상이 극히 단순한 경우에 국한되므로 유한요소법(FEM), 경계요소법(BEM)과 같은 수치해석적인 방법이 이용되고 있다. 경계요소법은 적분 커넬(kernel)의 특이성(singularity) 문제가 있지만 대상 영역의 경계면만을 이산화함으로써 모델링에 소요되는 시간과 노력을 절약할 수 있으므로 음향문제 해석에 있어서 효율적인 방법이라고 할 수 있다. 본 연구의 목적은 3차원 경계요소법 프로그램을 개발하고 평면파이론에 의한 해석이 어려운 여러가지 형태의 소음기에 대한 음향성능을 예측하고 실험으로 검증하는것이다. 특히, 단일영역으로 해석이 불가능한 다공형 소음기에 영역분할법을 적용하여 계산하고 결과를 검토하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6A
• /
• pp.575-583
• /
• 2002

JBIG2 is an International Standard fur compression of Bi-level images and documents. JBIG2 supports three encoding modes for high compression according to region features of documents. One of which is generic region coding for bitmap coding. The basic bitmap coder is either MMR or arithmetic coding. Pattern matching coding method is used for text region, and halftone pattern coding is used for halftone region. In this paper, a document is segmented into line-art, halftone and text region for JBIG2 encoding and JBIG2 CODEC is implemented. For efficient region segmentation of documents, region segmentation method using wavelet coefficient is applied with existing boundary extraction technique. In case of facsimile test image(IEEE-167a), there is improvement in compression ratio of about 2% and enhancement of subjective quality. Also, we propose arbitrary shape halftone region coding, which improves subjective quality in talc neighboring text of halftone region.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.1
• /
• pp.38-46
• /
• 2004

Image segmentation is the task which partitions the image into meaningful regions and considered to be one of the most important steps in computer vision and image processing. Image segmentation is also widely used in object-based video compression such as MPEG-4 to extract out the object regions from the given frame. Watershed algorithm is frequently used to obtain the more accurate region boundaries. But, it is well known that the watershed algorithm is extremely sensitive to gradient noise and usually results in oversegmentation. To solve such a problem, we propose an image segmentation method which is robust to noise by using stabilized inverse diffusion equation (SIDE) and is more efficient in segmentation by employing multiresolution approach. In this paper, we apply both the region projection method using labels of adjacent regions and the region merging method based on region adjacency graph (RAG). Experimental results on noisy image show that the oversegmenation is reduced and segmentation efficiency is increased.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.23 no.4
• /
• pp.369-378
• /
• 2010

The mixed finite element analysis is the most widely used method for saturated porous media. Generally, in this method, direct method and iterative method are proposed to obtain unknown variable, however, the iterative method is recommended because the method provide numerical stability and accuracy under the material properties for solid and fluid are different. In this paper, we introduce staggered method which has strong numerical stability, and FETI(Finite Element Tearing and Interconnecting) which is one of decomposition methods are applied into the method in order to obtain numerical efficiency. In which, Lagrange Multipliers and conjugated gradient method to solve decomposed domain are proposed, and then, the proposed method is verified numerical efficiency by point to point MPI(Message Passing Interface) library.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.693-696
• /
• 2005

한의학은 우수한 진단 능력을 보유하고 있음에도 불구하고 임상의의 직관에 머물러 있었던데 비해 서양의학은 다양한 진단 기기 개발로 인해 시장 성숙도 및 선호도에서 발전을 거듭하는 결과를 가져왔다. 본 논문에서는 한방의 직관적 진단법을 시각화, 객관화하기 위해 4진 방법 중 안면 망진 기법을 기기로 구현하기 위한 방법론을 제안하고자 한다. 이를 위해 우선적으로 찰색에 필요한 영역을 자동으로 분할하기 위한 방법을 개발하여 질병 진단에 필요한 진단 데이터를 시각화, 정량화, 계량화하기 위한 작업을 행하고자 한다. 또한 얼굴 안면의 색상 보정을 위해 화이트 밸런스 촬영으로 이미지를 분석하고 얼굴영역을 추출하며, 이목구비와 명당 영역을 추출하여 찰색에 필요한 영역을 분할해 내고자 한다. 끝으로 실험에 의해 제안한 방법의 유용성을 입증하고자 한다.