• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(5)
• /
• pp.131-133
• /
• 2000

In this paper, we proposed a new method for localizing the independent sources generating the observed EEG based on independent component analysis (ICA). The performance of the algorithm was tested through computer simulations.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2004년도 가을 학술발표논문집 Vol.31 No.2 (2)
• /
• pp.250-252
• /
• 2004

In this paper we apply several variations of independent component analysis( ICA) methods, such as spatial ICA (sICA), temporal ICA (tICA), and spatiotemporal ICA (stICA), to yeast cell cycle datasets, and compare their performance in finding components that result in gene clusters coherent with annotations and in extract ins meaningful temporal modes. It turns out that the results of tICA are superior to those of PCA, sICA, and stICA in terms of gene clustering and the temporal modes extracted by stICA highlights particular cellular processes.

• 한국해양공학회지
• /
• 제26권4호
• /
• pp.70-76
• /
• 2012

The independent component analysis (ICA) technique is a source identification method that uses statistical independence to separate source signals from measured signals. It has been successfully applied to various fields such as medical care and communication. In this study, the ICA technique was adopted to analyze the vibration source contribution of plate structures. The theory of the ICA technique is introduced and the procedure of the vibration source contribution analysis based on the ICA technique is proposed. To investigate the applicability of the proposed method to plate structures, numerical examples are presented for a rectangular plate under harmonic force excitations. The results show that the proposed method could become an effective tool for the vibration source contribution analysis of a plate structure.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.485-487
• /
• 2003

Independent Component Analysis (ICA) is used to generate ICA filter for computing feature vector for image window. Filters that have high discrimination power are selected to classify image from these ICA filters. Proposed classification algorithm is based on probability distribution of feature vector.

• 한국지능시스템학회논문지
• /
• 제13권3호
• /
• pp.371-376
• /
• 2003

본 논문에서는 효과적인 정보를 표현하는 Independent Component Analysis(ICA)-factorial 표현방법을 이용하여 얼굴감정 인식을 수행한다. 얼굴감정인식은 두 단계인 특징추출 과정과 인식과정에 의해 이루어진다. 먼저 특징추출방법은 주성분 분석(Principal Component Analysis)을 이용하여 얼굴영상의 고차원 공간을 저차원 특징공간으로 변환한 후 ICA-factorial 표현방법을 통해 좀 더 효과적으로 특징벡터를 추출한다. 인식단계는 최소거리 분류방법인 유클리디안 거리에 근거한 K-Nearest Neighbor 알고리즘으로 얼굴감정을 인식한다. 6개의 기본감정(기쁨, 슬픔, 화남, 놀람, 공포, 혐오)에 대해 얼굴 감정 데이터베이스를 구축하고 실험해본 결과 기존의 방법보다 좋은 인식 성능을 얻었다.

• Investigative Magnetic Resonance Imaging
• /
• 제3권2호
• /
• pp.188-195
• /
• 1999

fMRI의 신호는 매우 다양한 종류의 선호들이 혼합된 상태이며 , 비록 몇 가지의 요소에 대해 모델링하여 그 선호 형태를 추측할 수 있으나 모든 신호를 정확하게 분리하여 뇌신경의 활성화를 반영하는 신호만을 선택적으로 알아 내기는 어려운 일이다. 또한 뇌와 신체의 생리적 현상으로 발생하는 잡음뿐아니라 움직임이나 계기의 잡음은 fMRl의 데이터 분석을더욱 어렵게 한다. 따라서 실제 뇌신경의 활성화를 정확히 나타내는 참고데이터(reference data)를 선택하는 것은 힘든 일이며, 뇌신경의 활성화를 반영하는 의미 있는 여러 신호 형태에 대한 분석은 현재 fMRl의 후처리 (post-processing) 분석 방법에서 하나의 연구 과제라 할 수 있다. 본 연구에서는 prioriknow­-ledge 혹은 참고 데이터가 필요 없는 분석 방법인 Independent Component Analysis (lCA) 를 이용하여 fMRI선호를 분석하였다. ICA는 현재 많이 사용되고 있는 상관 분석 방법에 비해 신호의 형태를 분석하는 데에 보다 효과적일 수 있으며, 지연된 반응 형태를 갖는 신호나 움직임에 의한 신호의 패턴을 분리하여 분석할 수 있다. 한편, ICA만으후 fMRl의 신호에 따라 분석이 효과적이지 못한 경우 Principal Component Analysis(PCA) threshold, wavelet spatial f filtering, 부분적 영상 분석 방법들을 ICA전에 수행 함으로써 보다 효과적인 분석을 수행할 수 있다. ICA는 fMRl 신호의 형태 분석에 효과적인 방법이라고 생각하며, 데이터의 자유도를 감소 하기 위해서는 선 필터링 (pre-filtering) 방법들이 적용될 수 있다.

• Annals of Clinical Neurophysiology
• /
• 제3권1호
• /
• pp.26-30
• /
• 2001

Purpose : Eyeball movement is one of the main artifacts in EEG. A new approach to the removal of these artifacts is presented using independent component analysis(ICA). This technique is a signal-processing algorithm to separate independent sources from unknown mixed signals. This study was performed to show that ICA is a useful method for the separation of EEG components with little data deformity. Methods : 12 sets of 10 sec digital EEG data including eye opening and closure were obtained using international 10~20 system scalp electrodes. ICA with 18 tracings of double banana bipolar montage was performed. Among obtained 18 independent components, two components, which were thought to be eyeball movements were removed. Other 16 components were reconstructed into original bipolar montage. Power spectral analysis of EEGs before and after ICA was done and compared statistically. Total 12 pairs of data were compared by visual inspection and relative power comparison. Results : Waveforms of each pair looked alike by visual inspection. Means of relative power before and after ICA were 29.16% vs. 28.27%, 12.12% vs. 12.41%, 10.55% vs. 10.52%, and 19.33% vs. 18. 33% for alpha, beta, theta, and delta, respectively. These values were statistically same before and after ICA. Conclusions : We found little data deformity after ICA and it was possible to isolate eyeball movements in EEG recordings. Many other components of EEG could be selectively separated using ICA.

• 한국지능시스템학회논문지
• /
• 제14권5호
• /
• pp.623-629
• /
• 2004

In this paper, a Efficient method for the mixed image separation is presented using independent component analysis and the new fast expectation-maximization(EM) algorithm. In general, the independent component analysis (ICA) is one of the widely used statistical signal processing scheme in various applications. However, it has been known that ICA does not establish good performance in source separation by itself. So, Innovation process which is one of the methods that were employed in image separation using ICA, which produces improved the mixed image separation. Unfortunately, the innovation process needs long processing time compared with ICA or EM. Thus, in order to overcome this limitation, we proposed new method which combined ICA with the New fast EM algorithm instead of using the innovation process. Proposed method improves the performance and reduces the total processing time for the Image separation. We compared our proposed method with ICA combined with innovation process. The experimental results show the effectiveness of the proposed method by applying it to image separation problems.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.23-28
• /
• 2002

Among effective feature extraction methods from the small-patched image set, independent component analysis (ICA) is recently well known stochastic manner to find informative basis images. The ICA simultaneously learns both basis images and independent components using high order statistic manners, because that information underlying between pixels are sensitive to high-order statistic models. The topographic ICA model is adapted in our experiment. This paper deals with an unsupervised classification strategies using learned ICA basis images. The experimental result by proposed classification technique shows superior performance than classic texture analysis techniques for the panchromatic KOMPSAT imagery.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2006년도 춘계 종합학술대회 논문집
• /
• pp.381-384
• /
• 2006

본 논문에서는 독립성분분석(Independent Component Analysis, 이하 ICA)기법과 median에서의 GBD 알고리즘을 이용한 영상 분류 방법을 제안한다. 이 제시된 방법은 영상이 급격히 변화하는 부분의 정보를 잃지 않게 하면서 영상간의 거리를 측정할 수 있었다. 모의 실험 결과로부터 ICA는 K가 7에서 영상간의 상대적 식별이 불가능 하였지만, median에서는 K에 관계없이 영상간의 상대적 식별이 가능함을 모의 실험을 통하여 확인하였다.