• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.405-405
• /
• 2000

We designed the artificial olfactory sensing system (Electronic Nose) using MOS type sensor array fur recognizing and analyzing odour. The response of individual sensors of sensor array, each processing a slightly different response towards the sample volatiles, can provide enough information to discriminate between sample odours. In this paper, we applied clustering algorithm for dimension reduction, such as linear projection mapping (PCA method), nonlinear mapping (Sammon mapping method) and the combination of PCA and Sammon mapping having a better discriminating ability. The odours used are VOC (Volatile chemical compound) and Toxic gases.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.689-690
• /
• 2006

Relational discriminant analysis is a way of representing an object based on the dissimilarity measures among the prototypes extracted from feature vectors instead of the vectors themselves. Thus, by appropriately selecting a few number of representatives and by defining the dissimilarity measure, in this paper we propose a method of reducing the dimensionality and getting to achieve a better classification performance in both speed and accuracy. Our experimental results demonstrate that the proposed mechanism increases the performance as compared with the conventional approaches for samples involving artificial data sets.

• Communications for Statistical Applications and Methods
• /
• 제29권2호
• /
• pp.203-223
• /
• 2022

We propose a Bayesian approach to cumulative logistic regression model for the ordinal response based on the orthogonal principal components via singular value decomposition considering the multicollinearity among predictors. The advantage of the suggested method is considering dimension reduction and parameter estimation simultaneously. To evaluate the performance of the proposed model we conduct a simulation study with considering a high-dimensional and highly correlated explanatory matrix. Also, we fit the suggested method to a real data concerning sprout- and scab-damaged kernels of wheat and compare it to EM based proportional-odds logistic regression model. Compared to EM based methods, we argue that the proposed model works better for the highly correlated high-dimensional data with providing parameter estimates and provides good predictions.

• 한국음향학회지
• /
• 제22권1호
• /
• pp.35-41
• /
• 2003

일반적인 음성인식 방법에서는 주파수 전대역에서 추출한 특징벡터를 사용하므로, 각 주파수 부대역은 최종인식 결과에 동등하게 기여한다. 본 논문에서는 주파수 부대역별로 독립적인 특징을 추출하고, 음성인식에 효과적이 되도록 부대역의 켑스트럼 해상도를 조절하는 방법을 제안한다. 주파수 부대역별로 독립적인 특징을 추출하는 멀티밴드 음성인식접근을 사용하여 부대역 특징벡터의 차원을 변화시킨다. 최적의 벡터 차원 조합을 찾기 위하여 음성인식률과 군집화 품질을 사용한다. TIDIGITS 연결 숫자음을 사용한 실험결과에서, 제안한 방법은 전대역 특징추출에 비해 적은 계산량으로도 숫자열 인식률은 99.12%, 백분율 정확도 (percent correct)는 99.775%, 백분율 정밀도 (percent accuracy)는 99.705%를 얻었으며, 이는 전대역 특징벡터에 비해 상대적 오류율을 각각 38%, 32%, 37% 감소시킨 결과이다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권1호
• /
• pp.65-70
• /
• 2013

최근 디지털 카메라 기술의 발전으로 이미지를 쉽게 생성할 수 있어 이를 활용한 컴퓨터 비전분야의 연구가 활발하게 진행되고 있다. 특히 디지털 이미지에서 특징점을 추출하고 이를 활용하는 연구가 활발하게 진행되고 있다. 이미지 스티칭은 여러 이미지에서 특징점을 추출하고 이를 정합하여 하나의 고해상도 이미지를 생성하는 것으로 군사용, 의료용뿐만 아니라 실생활의 다양한 분야에서 활용되고 있다. 본 논문에서는 특징점 기술자의 차원을 효과적으로 감소시켜 정확하면서도 빠르게 정합점을 찾을 수 있는 SURF 기반의 빠른 특징점 기술자 추출 및 정합을 이용한 효율적인 이미지 스티칭 기법을 제안한다. 추출된 특징점에서 불필요한 특징점을 분류하여 특징점 기술자를 생성한다. 이때 특징점 기술자의 연산량을 줄이면서도 효율적인 정합을 위해 기술자의 차원을 줄이고 방향 윈도우를 확장하였다. 실험 결과 특징점 정합 및 전체 이미지 스티칭 속도가 기존의 알고리즘보다 빠르면서도 자연스러운 스티칭된 이미지를 생성할 수 있었다.

• 말소리와 음성과학
• /
• 제1권4호
• /
• pp.127-132
• /
• 2009

The feature vectors which are used in conventional speaker recognition (SR) systems may have many correlations between their neighbors. To improve the performance of the SR, many researchers adopted linear transformation method like principal component analysis (PCA). In general, the linear transformation of the feature vectors is based on concatenated form of the static features and their dynamic features. However, the linear transformation which based on both the static features and their dynamic features is more complex than that based on the static features alone due to the high order of the features. To overcome these problems, we propose an efficient method that applies linear transformation and temporal information of the features to reduce complexity and improve the performance in speaker verification (SV). The proposed method first performs a linear transformation by PCA coefficients. The delta parameters for temporal information are then obtained from the transformed features. The proposed method only requires 1/4 in the size of the covariance matrix compared with adding the static and their dynamic features for PCA coefficients. Also, the delta parameters are extracted from the linearly transformed features after the reduction of dimension in the static features. Compared with the PCA and conventional methods in terms of equal error rate (EER) in SV, the proposed method shows better performance while requiring less storage space and complexity.

• International Journal of Precision Engineering and Manufacturing
• /
• 제2권2호
• /
• pp.65-72
• /
• 2001

The design of multi-stage gear drives is a time-consuming process, since on includes more complicated problems, which are not considered in the design of single-stage gear drives. The designer has th determine the number of reduction stages and the gear ratios of each reduction state. In addition, the design problems include not only the dimensional design but also the configuration design of gear drive elements. There is no definite rule and principle for these types of design problems. Thus the design practices largely depend on the sense and the experiences of the designer , and consequently result in undesirable design solution. We propose a new generalized design algorithm to support the designer at the preliminary design phase of multi-stage gear drives. The proposed design algorithm automates the design process by integrating the dimensional design and the configuration design process. The algorithm consists of four steps. In the first step, a designer determines the number of reduction stage. In the second step. gear ratios se chosen by using the random search method. In the third step, the values of basic design parameter are chosen by using the generate and test method. Then, the values of other dimension, such ad pitch diameter, outer diameter, and face width, are calculated for the configuration design in the final step. The strength and durability of a gear is guaranteed by the bending strength and the pitting resistance rating practices by using the AGMA rating formulas. In the final step, the configuration design is carried out b using the simulated annealing algorithm. The positions of gears and shafts are determined to minimize the geometrical volume(size) of a gearbox, while satisfying spatial constraints between them. These steps are carried out iteratively until a desirable solution is acquired. The propose design algorithm has been applied to the preliminary design of four-stage gear drives in order to validate the availability. The design solution have shown considerably good results in both aspects of the dimensional and the configuration design.

• 디지털융복합연구
• /
• 제11권8호
• /
• pp.183-190
• /
• 2013

PPG 신호는 심장의 박동에 동기된 유사 주기 신호이다. 본 논문에서는 PPG 신호의 유사주기성을 이용한 지수가중 이동평균필터 방법을 제안한다. 이 필터링 방법은 PPG 신호를 주기적으로 분리하여 각 주기 신호의 같은 순번에 있는 샘플들끼리 평균을 취하는 방법이다. 연속된 PPG 신호의 주기중에 동잡음이 혼입되었다면 주기를 기준으로 PPG 신호를 분리한 후, 각 주기의 샘플수를 조정하여 같은 샘플수를 가지게 만든다. 이 주기들을 2차원으로 배열한 후 현재 주기부터 이전 각 주기의 샘플끼리 평균을 취함으로써 훼손없이 동잡음을 제거할 수 있었다.

• Structural Engineering and Mechanics
• /
• 제72권6호
• /
• pp.763-774
• /
• 2019

The main aim of this paper is enhancing design of traditional laminated composite plates subjected to static loads. In this regard, this paper suggests embedding a lightweight porous layer in the middle of laminated composite as the core layer of the resulted sandwich plate. The static responses of the suggested structures with uniform, symmetric and non-symmetric porosity distributions are compared to optimize their design. Using the first order shear deformation theories, the static governing equations of the suggested laminated composite plates with a porous layer (LCPPL) rested on two-parameter foundation are obtained. A finite element method is also utilized to solve the governing equations of LCPPLs. Effects of laminated composite and porosity characteristics as well as geometry dimension, edges' boundary conditions and foundation coefficients on the static deflection and stress distribution of the suggested composite plates have been investigated. The results reveal that the use of core between the layers of laminated composites leads to a sharp reduction in the static deflections of LCPPLs. Furthermore, in compare with perfect cores, the use of porous core between the layers of laminated composite plates can offer a considerable reduction in structural weight without a significant difference in their static responses.

• 한국컴퓨터정보학회논문지
• /
• 제17권4호
• /
• pp.75-82
• /
• 2012

PPG 신호는 심장의 박동에 동기된 유사 주기 신호이다. 본 논문에서는 PPG 신호의 유사주기성을 이용한 주기적 이동평균필터 방법을 제안한다. 이 필터링 방법은 PPG 신호를 주기적으로 분리하여 각 주기 신호의 같은 순번에 있는 샘플들끼리 평균을 취하는 방법이다. 연속된 PPG 신호의 주기 중에 동잡음이 혼입되었다면 주기를 기준으로 PPG 신호를 분리한 후, 각 주기의 샘플수를 조정하여 같은 샘플수를 가지게 만든다. 이 주기들을 2차원으로 배열한 후 현재 주기부터 이전 각 주기의 샘플끼리 평균을 취함으로써 훼손 없이 동잡음을 제거할 수 있었다.