• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.2
• /
• pp.99-107
• /
• 2009

This paper presents an enhanced hybrid model based on Independent Component Analysis(ICA) in order to features of numeric characters in license plates. ICA which is used only in high dimensional statistical features doesn't consider statistical features in low dimension and correlation between numeric characters. To overcome the drawbacks of ICA, we propose an improved ICA with the hybrid model using both Principle Component Analysis(PCA) and Linear Discriminant Analysis(LDA). Experiment results show that the proposed model has a superior performance in feature extraction and recognition compared with ICA only as well as other hybrid models.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04a
• /
• pp.297-300
• /
• 2001

뇌파(Electroencephalogram, EEG)는 뇌의 자발적 전기활동을 두피에서 측정한 것이다. 그 동안 뇌질환과 관련된 임상에서 주로 사용되어져 왔으며, 비선형 동역학 연구를 통해 결정론적인 동역학 신호임이 밝혀짐에 따라 뇌 기능연구 분야에서 그 응용범위가 넓어지고 있다. 우리는 뇌파 신호에 대하여 독립성분분석(Independent Component Analysis, ICA)을 통하여 그 결과를 알아보았다. 즉, 뇌파의 독립성분 분석 적용 타당성을 알아본 다음 이를 적용하여 독립 소스들을 분리해 내었다. 또한 Topological Mapping을 이용하여 각각의 독립 소스들이 뇌의 어느 위치에서 발생하는지도 알아보았다. 이를 통하여 EEG에 독립성분분석을 적용함으로써 뇌 활동의 시간적, 공간적 분석이 가능하고 유용함을 나타내었다.

• Progress in Medical Physics
• /
• v.18 no.1
• /
• pp.20-26
• /
• 2007

It is expected that synaptic construction and electrical characteristics In degenerate retina might be different from those In normal retina. Therefore, we analyzed the retinal waveform recorded with multielectrode array in normal and degenerate retina using principal component analysis (PCA) and Independent component analysis (ICA) and compared the results. PCA Is a well established method for retinal waveform while ICA has not tried for retinal waveform analysis. We programmed ICA toolbox for spatiotemporal analysis of retinal waveform. In normal mouse, the MEA spatial map shows a single hot spot perfectly matched with PCA-derived ON or OFF ganglion cell response. However In rd/rd mouse, the MEA spatial map shows numerous hot and cold spots whose underlying interactions and mechanisms need further Investigation for better understanding.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.5
• /
• pp.442-450
• /
• 2002

In measuring EEG, which is widely used for studying brain function, EEG is frequently mixed with noise and artifact. In this study, the signals relevant to the artifact were distracted by applying ICA to EEG signal. First, each independent component which was assumed to be the source was separated by applying ICA to EEG which involved artifact relevant to the eye movement of a normal person. Next, the signal which was assumed to be artifact was removed from the separated 18 independent components, and the nonlinear analysis method such as correlation dimension and the Iyapunov exponent was applied to each reconstructed EEG signal and the original signal including artifact in order to find meaningful difference between the two signals and infer the anatomical localization of its source and distribution. This study shows it is possible not only to analyze the brain function visually and spatially for visually complex EEG signal, but also to observe its meaningful change through the quantitative analysis of EEG by means of the nonlinear analysis.

• The Korean Journal of Applied Statistics
• /
• v.19 no.1
• /
• pp.135-148
• /
• 2006

Independent component analysis (ICA) is a statistical method for transforming an observed high-dimensional multivariate data into statistically independent components. ICA has been applied increasingly in wide fields of spectrum application since ICA is able to extract unknown components of a mixture from spectra. We focus on application of ICA for separating independent sources and predicting each composition using extracted components. The theory of ICA is introduced and an application to a metal surface spectra data will be described, where subsequent analysis using non-negative least square method is performed to predict composition ratio of each sample. Furthermore, some simulation experiments are performed to demonstrate the performance of the proposed approach.

• Journal of KIISE:Software and Applications
• /
• v.27 no.5
• /
• pp.583-592
• /
• 2000

This paper proposes feature parameter transformation method using independent component analysis (ICA) for speaker identification. The proposed method assumes that the cepstral vectors from various channel-conditioned speech are constructed by a linear combination of some characteristic functions with random channel noise added, and transforms them into new vectors using ICA. The resultant vector space can give emphasis to the repetitive speaker information and suppress the random channel distortions. Experimental results show that the transformation method is effective for the improvement of speaker identification system.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.351-359
• /
• 2003

In this paper, we proposed a method for acoustic echo cancellation based on independent component analysis. When the large acoustic noise is picked up by the microphone, the performance of echo cancellation decreased. We used two microphones that received echo signal which is linearly mixed with the noise, then separated the echo signals from the received signals with independent component analysis algorithm. The separated echo signal is used for the reference signal of adaptive algorithm which leads to better performance of the echo cancellation. Computer simulation results show the validity of the proposed method.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.539-544
• /
• 2003

This paper describes how to reduce the effect of an occupation threshold by that the transform of mixture components of HMM parameters is controlled in hierarchical tree structure to prevent from over-adaptation. To reduce correlations between data elements and to remove elements with less variance, we employ PCA (Principal component analysis) and ICA (independent component analysis) that would give as good a representation as possible, and decline the effect of over-adaptation. When we set lower occupation threshold and increase the number of transformation function, ordinary MLLR adaptation algorithm represents lower recognition rate than SI models, whereas the proposed MLLR adaptation algorithm represents the improvement of over 2% for the word recognition rate as compared to performance of SI models.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.3
• /
• pp.1319-1324
• /
• 2012

Tomatoes were harvested at different ripening stages. To determine the ripening stages, We analyzed the relation between the compound concentrations of tomato measured with HPLC and the tomato RGB images. Among the compound concentrations, tomato quality is mostly affected by the Lycopene. The $Q^2$ error of the predicted Lycopene concentration and the corresponding independent component of tomato RGB image, determined from the PLS procedure, was 0.92. and we show the effectiveness of the independent component by comparing the error between the pixel area of RGB image applied by independent component and the simple black white tomato image. This regression made it possible to construct concentration images of the tomatoes, which showed non-uniform ripening. The method can be applied in an unsupervised real time sorting machine of unripe and discolored tomato using the compound concentrations.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.370-373
• /
• 2002

본 논문에서는 단일영상들에 포함된 특징들을 효과적으로 추출하기 위하여 신경망 기반 독립성분분석기법의 이용을 제안하였다. 여기서 독립성분의 효과적인 분석을 위해 고정점 학습알고리즘의 신경망 기반 기법을 이용하였다. 이는 수치적 기법에 비해 신경망이 가지는 ？ㄱ습 등의 우수한 속성과 뉴우턴법의 고정점 알고리즘이 가지는 빠르고 간단한 계산속성을 동시에 살리기 위함이다. 제안된 기법을 512x412 픽셀의 L둠 영상과 480x225 픽셀의 지폐영상 각각에서 선택된 1,000개의 영상패치들을 대상으로 시뮬레이션 한 결과, 추출된 16x16 펙셀의 160개 독립성분 기저벡터는 지문영상과 지폐영상 각각에 포함된 공간적인 주파수 특성과 방향성을 가지는 경계 특성이 잘 드러나는 국부적인 특징들임을 확인할 수 있었다.