http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
냄새인식을 위한 강력한 Back-propagation 알고리즘의 제안
변형기 三陟大學校 1998 論文集 Vol.31 No.1
Artificial neural networks are increasingly being used to enhance the classification and recognition powers of data collected from sensor array. This paper reports the effectiveness of multilayer perceptron network based on back-propagation algorithm combined with the output patterns from an artificial sensing system using eletrically conducting polymers as sensor materials. The proposed three layer architecture of a multilayer perceptron network, which has newly proposed pre-processing method and optimal rates of learning and momentum, produced robust performance and classification results throughout the experimental trails.
전자코 시스템을 사용한 휘발성 화학물질 분류를 위한 퍼지LVQ(Fuzzy Learning Vector Quantization)알고리즘의 응용
변형기 三陟大學校 1997 論文集 Vol.30 No.1
A pattern recognition technique based on fuzzy set theory is applied to gas and odour classification; fuzzy learning vector quantization(FLVQ). This is an algorithm which is derived from Kohonen's learning vector quantization(LVQ), and is modified and extended by fuzzy set theory. The training procedure of FLVQ is more to the position of the reference patterns and modified the fuzziness. The self-organization of FLVQ has been realized using this training method. The performance of FLVQ confirmed the classification of volatile chemicals. Moreover, it is possible to distinguish an unknown pattern from a known one utilizing the FLVQ; the experimental work confirmed this capability.
후각인지 시스템 데이타의 시각적 판별을 위한 Unsupervised 분류방법
김갑일,변형기 明知大學校 産業技術硏究所 1996 産業技術硏究所論文集 Vol.15 No.-
Display 방법은 일반적으로 패턴인식응용 분야의 첫번째 순서이다. Conducting PolYmer를 이용한 센서Array는 가스및냄새들을 측정, 감지된 가스및냄새에 따라 다차원(Multi-dimensional)의 서로 다른 패턴들을 보여준다. 측정된 가스및냄새들의 다차원패턴들을 이들이 갖고있는 특성변환 없이 저차원(Low-dimensional)패턴으로 바꾸면, 인간의 시각을 이용 쉽게 분류 할 수 있다. 본 논문 에서는, 다차원공간의 패턴을 인간인 인식하기 쉬운 2차원 패턴으로 줄이기위해 Karhunen-Loeve Expansion을 이용한 선형 Projection 알고리즘과 Sammon의 비선형 Mapping 알고리즘을 제안하였다. 특히, 각 cluster간의 패턴분포에 따른 Boundary Criteria는 통계학적 방법인 Chi-Square Distribution을 이용하여 정하여 졌다. 이같은 방법들은 Unsupervised를 기본으로 함으로서 Prior Assumption이 필요하지 않으며, Mapping후 단지 시각만을 이용하여 가스나 냄새를 쉽게 분류한다.
전도성 고분자 센서 어레이를 이용한 휘발성 유기 화합물 가스 인식
이경문,주병수,유준부,황하룡,이병수,이덕동,변형기,허증수 한국센서학회 2002 센서학회지 Vol.11 No.5
휘발성 유기 화합물 가스(Volatile Organic Compounds)를 인식하고 분석하기 위하여 전도성 고분자 센서어레이를 이용한 시스템을 제작하였다. Polypyrrole와 Polyaniline을 화학중합법으로 센서에 전도성고분자막을 형성하였고 이를 통해 VOC 검지용 센서 어레이를 제작하였다. 센서어레이로부터 측정되는 다차원 데이터는 주성분분석법(PCA)과 RBF(Radial Basis Function Network)을 이용하였다. 제안된 시스템으로 VOCs 가스를 인식하는데 있어서 RBF Network이 PCA방식보다 더욱 효율적인 것으로 판단되었다. We fabricated gas recognition system using conducting polymer sensor array for recognizing and analyzing VOCs(Volatile Organic Compounds) gases. The polypyrrole and polyaniline thin film sensors which were made by chemical polymerization were employed to detect VOCs. The multi-dimensional sensor signals obtained from the sensor array were analyzed using PCA(principal component analysis) technique and RBF(radial basis function) Network. Throughout the experimental trails, we confirmed that RBF Network is effective than PCA technique in identifying VOCs.
( Hyung Gi Byun ) 한국센서학회 2011 센서학회지 Vol.20 No.3
A variety of pattern recognition algorithms including neural networks may be applicable to the identification of odors. In this paper, an identification technique for an electronic odor sensing system applicable to wound state monitoring is presented. The performance of the radial basis function(RBF) network is highly dependent on the choice of centers and widths in basis function. For the fine tuning of centers and widths, those parameters are initialized by an ill-conditioned genetic fuzzy c-means algorithm, and the distribution of input patterns in the very first stage, the stochastic gradient(SG), is adapted. The adaptive RBF network with singular value decomposition(SVD), which provides additional adaptation capabilities to the RBF network, is used to process data from array-based gas sensors for early detection of wound infection in burn patients. The primary results indicate that infected patients can be distinguished from uninfected patients.
( Hyung-gi Byun ) 한국센서학회 2016 센서학회지 Vol.25 No.3
The use of a chemical sensor array can help discriminate between chemicals when comparing one samplewith another. The ability to classify pattern characteristics from relatively small pieces of information has led to growing interest in methods of sensor recognition. A variety of pattern recognition algorithms, including the adaptive radial basis function network(RBFN), may be applicable to gas and/or odor classification. In this paper, we provide a broad review of approaches for various types of gas and/or odor identification techniques based on RBFN and drift compensation techniques caused by sensor poisoning and aging.
Intelligent Electronic Nose System for Detection of VOCs in Exhaled Breath
( Hyung-gi Byun ),( Joon-bu Yu ) 한국센서학회 2019 센서학회지 Vol.28 No.1
Significant progress has been made recently in detection of highly sensitive volatile organic compounds (VOCs) using chemical sensors. Combined with the progress in design of micro sensors array and electronic nose systems, these advances enable new applications for detection of extremely low concentrations of breath-related VOCs. State of the art detection technology in turn enables commercial sensor systems for health care applications, with high detection sensitivity and small size, weight and power consumption characteristics. We have been developing an intelligent electronic nose system for detection of VOCs for healthcare breath analysis applications. This paper reviews our contribution to monitoring of respiratory diseases and to diabetic monitoring using an intelligent electronic nose system for detection of low concentration VOCs using breath analysis techniques.