• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.33-38
• /
• 2014

The FCM based hybrid RBF network is a heterogeneous learning network model that applies FCM algorithm between input and middle layer and applies Max_Min algorithm between middle layer and output. The Max-Min neural network uses winner nodes of the middle layer as input but shows inefficient learning in performance when the input vector consists of too many patterns. To overcome this problem, we propose a dynamic learning rate control based on fuzzy logic. The proposed method first classifies accurate/inaccurate class with respect to the difference between target value and output value with threshold and then fuzzy membership function and fuzzy decision logic is designed to control the learning rate dynamically. We apply this proposed RBF network to the character recognition problem and the efficacy of the proposed method is verified in the experiment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.5
• /
• pp.967-973
• /
• 2017

Clustering is one of the most well-known unsupervised learning methods that clusters data into homogeneous groups. Clustering has been used in various applications and FCM is one of the representative methods. In Fuzzy C-Means(FCM), however, cluster centers tend leaning to high density areas because the Euclidean distance measure forces high density clusters to make more contribution to clustering result. Previously proposed was density-independent clustering method, where cluster centers were made not to be close each other and relived the center deviation problem. Density-independent clustering method has a limitation that it is difficult to specify the position of the cluster centers. In this paper, an enhanced density-independent clustering method with an additional term that makes cluster centers to be placed around dense region is proposed. The proposed method converges more to real centers compared to FCM and density-independent clustering, which can be verified with experimental results.

• Journal of Intelligence and Information Systems
• /
• v.13 no.1
• /
• pp.81-90
• /
• 2007

This paper shows an intelligent disease diagnosis system for public. Our system deals with 30 diseases and their typical symptoms selected based on the report from Ministry of Health and Welfare, Korea. Technically, the system uses a modified FCM algorithm for clustering diseases and the input vector consists of the result of user-selected questionnaires. The modified FCM algorithm improves the quality of clusters by applying symmetrically measure based on the fuzzy theory so that the clusters are relatively sensitive to the shape of the pattern distribution. Furthermore, we extract the highest 5 diseases only related to the user-selected questionnaires based on the fuzzy membership function between questionnaires and diseases in order to avoid diagnosing unrelated disease.

• Food Science and Biotechnology
• /
• v.15 no.5
• /
• pp.739-745
• /
• 2006

Changes in the levels of analytes in the blood and urine of a rodent animal model were taken as a measure of the hypoglycemic effects of a diet containing fermented chaga mushroom. These studies were conducted using the genetically manipulated diabetic Otsuka Long-Evans Tokushima Fatty (OLETF) rat. The effects of 8-week long diets that included either fermented (FCM) or non-fermented (CM) chaga mushroom powder (5% in the diet) on the OLETF rat were compared to the normal diet fed OLETF rat and the non-diabetic Long-Evans Tokushima Otsuka (LETO) rat. Hypoglycemia was tracked by measuring serum and urine concentrations of glucose, insulin, fructosamine, and leptin. Serum and urine levels of glucose, fructosamine, and leptin in the OLETF rats were higher than in LETO rats when fed normal diets but insulin levels did not differ between the two animal groups. The FCM rats were characterized by dramatically low levels of serum glucose and leptin in the OLETF rats whereas the levels of fructosamine and urine glucose trended lower in response to FCM. The serum leptin level in the CM-fed OLETF rat was also lower than that in the normal diet fed OLETF control. Serum concentrations of insulin in the OLETF rats were higher following FCM or CM feeding compared to the normal diet. These observations imply that (a) a dietary supplement of fermented chaga mushroom may contribute to a hypoglycemic effect in the OLETF rat, and (b) the increased blood insulin concentration following 8 weeks of an FCM diet may be important to the noted improvement in hyperglycemia.

• Environmental Engineering Research
• /
• v.17 no.1
• /
• pp.35-39
• /
• 2012

In this study, multiparametric flow cytometry (FCM) was installed to enumerate the diagnosis of Pseudomonas aeruginosa ATCC 10145 and Escherichia coli K12 (IFO 3301). The nucleic acids (DNA/RNA) were double stained by a LIVE/DEAD bacLight viability kit, involving green SYTO 9 and red propidium iodide (PI), based on the permeability of two chemicals according to the integrity of plasma membrane. As the results showed, the gate for dead bacteria was defined as the range of $0.2{\times}10^0$ to $6.0{\times}10^1$ photo multiplier tube (PMT) 2 fluorescence (X-axis) and $2.0{\times}10^0$ to $2.0{\times}10^2$ PMT 4 fluorescence (Y-axis), and the gate for live bacteria was defined as the range of $6.0{\times}10^0$ to $6.0{\times}10^2$ PMT 2 fluorescence (X-axis) and $2.0{\times}10^0$ to $4.0{\times}10^2$ PMT 4 fluorescence (Y-axis). In the comparison of the number of the tested bacteria detected by FCM (viability assessment) and plate culture (cultivability assessment), the number of bacteria detected by FCM well represented the number of bacteria that was detected by the colony forming unit (CFU) counting method when bacteria were exposed to isopropyl alcohol and silver/copper cations. Consequently, it is concluded that the application of FCM to monitor the functional effect of disinfectants on the physiological status of target bacteria can offer more rapid and reliable data than the plate culture colony counting method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.71-73
• /
• 2017

본 논문에서는 4 방향 윤곽선 추적 기법과 히스토그램 분석 기법을 기반으로 한 개선된 FCM 알고리즘을 적용하여 색조 도플러 초음파 영상에서 상완 동맥의 혈류를 추출하고 분석하는 방법을 제안한다. 제안된 방법에서는 상완 동맥의 혈류를 정확히 추출하기 위해 전처리 과정으로 색조 도플러 초음파 영상 이외의 환자 정보가 있는 영역을 제거한 후, ROI 영역을 추출한다. 추출된 ROI 영역에서 영상의 최대 명암도를 임계치로 설정한 이진화 기법을 적용하여 ROI 영역을 이진화한다. 이진화된 ROI 영역에서 4 방향 윤곽선 추적 기법을 적용하여 상완 동맥이 존재하는 사다리꼴 형태의 영역을 추출한다. 색 정보를 분석한 히스토그램을 이용하여 특징점의 개수를 계산하고 계산된 특징점의 개수를 FCM 알고리즘의 초기 클러스터의 개수로 설정한 후, 추출된 사다리꼴 형태의 영역에 적용하여 양자화 한다. 양자화된 영역 중에서 빨간색으로 분류된 영역을 고혈압 영역으로 추출한다. 제안된 추출 방법을 20개의 색조 도플러 초음파 영상을 대상으로 실험한 결과, 20개의 색조 도플러 초음파 영상에서 18개의 색조 도플러 초음파 영상이 정확히 추출되었다.

• Nuclear Engineering and Technology
• /
• v.48 no.3
• /
• pp.650-659
• /
• 2016

As a type of accident-tolerant fuel, fully ceramic microencapsulated (FCM) fuel was proposed after the Fukushima accident in Japan. The FCM fuel consists of tristructural isotropic particles randomly dispersed in a silicon carbide (SiC) matrix. For a fuel element with such high heterogeneity, we have proposed a two-temperature homogenized model using the particle transport Monte Carlo method for the heat conduction problem. This model distinguishes between fuel-kernel and SiC matrix temperatures. Moreover, the obtained temperature profiles are more realistic than those of other models. In Part I of the paper, homogenized parameters for the FCM fuel in which tristructural isotropic particles are randomly dispersed in the fine lattice stochastic structure are obtained by (1) matching steady-state analytic solutions of the model with the results of particle transport Monte Carlo method for heat conduction problems, and (2) preserving total enthalpies in fuel kernels and SiC matrix. The homogenized parameters have two desirable properties: (1) they are insensitive to boundary conditions such as coolant bulk temperatures and thickness of cladding, and (2) they are independent of operating power density. By performing the Monte Carlo calculations with the temperature-dependent thermal properties of the constituent materials of the FCM fuel, temperature-dependent homogenized parameters are obtained.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.1347-1348
• /
• 2015

본 논문에서는 기상레이더 데이터에서 섞여있는 강수에코 및 비강수에코를 분류하기 위하여 Interval Type-2 FCM based RBFNN의 도움으로 사례 및 에코 분류기의 설계를 제안한다. 학습과 테스트 데이터는 현재 기상청에서 사용하는 UF radar data를 사용하였으며, 사례 분류기와 에코패턴 분류기의 데이터를 각각 생성한다. 전처리 과정인 사례 분류를 통하여 강수사례 혹은 비강수사례를 분류하여 강수사례일 경우 에코패턴분류를 진행하며, 비강수사례일 경우 데이터에 관측된 모든 반사도 값을 제거한다. 사례 및 에코 분류기는 Interval Type-2 FCM based RBFNN을 통하여 패턴분류를 진행하며, 패턴분류 성능을 확인한다. 또한 후반부 파라미터의 동정 시, 각 규칙에 파라미터를 전역적으로 구하는 LSE와 각 규칙에 대한 파라미터를 독립적으로 구하는 WSLE의 비교연구를 수행한다. 분류기의 성능을 확인하기 위하여 사례 분류 후 에코패턴분류의 결과는 현재 기상청에서 사용하고는 품질검사(QC) 데이터와 비교하여 평가하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.252-257
• /
• 2007

우리나라의 항만은 수 출입화물의 99.5%를 처리하며, 육로 및 철도 수송 물동량의 기종점 역할을 수행하는 중요한 곳으로서 항만 물동량의 신속한 처리와 자동화 시스템에 의한 비용절감은 엄청난 효과를 가져온다. 따라서 본 논문에서는 항만에서 취급하는 컨테이너를 자동으로 식별할 수 있는 자동화 방법을 제안한다. 실제 컨테이너 영상을 그레이 영상으로 변환한 후, 프리윗 마스크(Prewitt-Mask)를 적용하여 윤곽선을 추출하고 컨테이너를 식별할 수 있는 개별 식별자의 형태학적 특징 정보를 이용하여 식별자 후보영역을 검출한다. 검출된 식별자 후보영역은 개별 식별자 영역외에 잡음 영역이 포함되어 있으므로 4방향 윤곽선 추적 알고리즘과 Grassfire 알고리즘을 적용하여 잡음을 제거하고 개별 식별자들을 각각 객체화한다. 잡음이 제거된 식별자 후보 영역에서 객체화 한 개별 식별자는 컨테이너 식별을 위해 FCM 기반 퍼지 RBF 네트워크를 적용하여 인식한다. 본 논문에서 제안한 컨테이너 식별자 인식 방법의 성능을 평가하기 위해 실제 컨테이너 영상 300장을 대상으로 실험한 결과, 기존의 방법보다 인식 성능이 개선되었음을 확인할 수 있었다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.46 no.12
• /
• pp.102-107
• /
• 2009

As technology gets developed, medical equipments are also modernized and leading-edge systems, such as PACS become popular. Many scientists noticed importance of medical image processing technology. Technique of region segmentation is the first step of digital medical image processing. Segmentation technique helps doctors to find out abnormal symptoms early, such as tumors, edema, and necrotic tissue, and helps to diagnoses correctly. Segmentation of white matter, gray matter and CSF of a brain image is very crucial part. However, the segmentation is not easy due to ambiguous boundaries and inhomogeneous physical characteristics. The rate of incorrect segmentation is high because of these difficulties. Fuzzy-based segmentation algorithms are robust to even ambiguous boundaries. In this paper a modified Fuzzy-based segmentation algorithm is proposed to handle the noise of MR scanners. A proposed algorithm requires minimal computations of mean and variance of neighbor pixels to adjust a new neighbor list. With the addition of minimal compuation, the modified FCM(mFCM) lowers the rate of incorrect clustering below 30% approximately compared the traditional FCM.