Interval Type-2 Fuzzy C-means Research Articles

적응적 Multiple Kernels을 이용한 Interval Type-2 Possibilistic Fuzzy C-Means 방법

본 논문에서는 interval type-2 possibilistic fuzzy C-means(IT2PFCM) 클러스터링 방법에 multiple Gaussian kernels을 기반으로 한 possibilistic fuzzy C-means multiple kernels(PFCM-MK) 알고리즘을 결합하여 적응적인 하이브리드 클러스터링 방법인 multiple kernels interval type-2 possibilistic fuzzy C-means(IT2PFCM-MK) 방법을 제안 하였다. 일반적으로 possibilistic fuzzy C-means(PFCM) 알고리즘은 fuzzy C-means(FCM) 알고리즘의 단점인 노이즈 민감성 및 특이점 문제와 알고리즘 초기 클러스터의 Prototype에 따라 위치가 겹치는 문제를 해결하기 위해 제안 되었다. 하지만 이 방법 역시 퍼지화 파라미터 값에 따라 위와 같은 문제를 여전히 가지고 있기 때문에 이와 같은 문제를 보완하기 위해 interval type-2 퍼지 접근 방법을 이용 하는 interval type-2 possibilistic fuzzy C-means(IT2PFCM) 알고리즘을 제안 하였다. 또한 multiple kernels 함수를 interval type-2 possibilistic fuzzy C-means(IT2PFCM) 알고리즘에 적용하여 분류하기 복잡한 형태의 데이터와 노이즈가 있는 데이터에 대하여 보다 정확하고, 향상된 클러스터링을 수행할 수 있다. In this paper, we propose a hybrid approach towards multiple kernels interval type-2 possibilistic fuzzy C-means(PFCM) based on interval type-2 possibilistic fuzzy c-means(IT2PFCM) and possibilistic fuzzy c-means using multiple kernels( PFCM-MK). In case of noisy data or overlapping cluster prototypes, fuzzy C-means gives poor performance in comparison to possibilistic fuzzy C-means(PFCM). Moreover, to address the uncertainty associated with fuzzifier parameter m, interval type-2 possibilistic fuzzy C-means(PFCM) is used. Most of the practical data available are complex and non-linearly separable. In such cases using Gaussian kernels proves helpful. Therefore, in order to overcome all these issues, we have integrated multiple kernels possibilistic fuzzy C-means(PFCM) into interval type-2 possibilistic fuzzy C-means(IT2PFCM) and propose the idea of multiple kernels based interval type-2 possibilistic fuzzy C-means(IT2PFCM-MK).

A genetic type-2 fuzzy C-means clustering approach to M-FISH segmentation

Multiplex Fluorescent In Situ Hybridization (M-FISH) is a multi-channel chromosome image generating technique that allows colors of the human chromosomes to be distinguished. In this technique, all chromosomes are labelled with 5 fluors and a fluorescent DNA stain called DAPI (4 in, 6-Diamidino-2-phenylindole) that attaches to DNA and labels all chromosomes. Therefore, a M-FISH image consists of 6 images, and each image is the response of the chromosome to a particular fluor. In this paper, we propose a genetic interval type-2 fuzzy c-means (GIT2FCM) algorithm, which is developed and applied to the segmentation and classification of M-FISH images. Chromosome pixels from the DAPI channel are segmented by GIT2FCM into two clusters, and these chromosome pixels are used as a mask for the remaining five channels. Then, the GIT2FCM algorithm is applied to classify the chromosome pixels into 24 classes, which correspond to the 22 pairs of homologous chromosomes and two sexual chromosomes. The experiments performed using the M-FISH dataset show the advantages of the proposed algorithm.

Enhanced interval type-2 fuzzy c-means algorithm with improved initial center

Uncertainties are common in the applications like pattern recognition, image processing, etc., while FCM algorithm is widely employed in such applications. However, FCM is not quite efficient to handle the uncertainties well. Interval type-2 fuzzy theory has been incorporated into FCM to improve the ability for handling uncertainties of these algorithms, but the complexity of algorithm will increase accordingly. In this paper an enhanced interval type-2 FCM algorithm is proposed in order to reduce these shortfalls. The initialization of cluster center and the process of type-reduction are optimized in this algorithm, which greatly reduce the calculation time of interval type-2 FCM and accelerate the convergence of the algorithm. Many simulations have been performed on random data clustering and image segmentation to show the validity of our proposed algorithm.

A modified interval type-2 fuzzy C-means algorithm with application in MR image segmentation

The fuzzy C-means (FCM) algorithm has significant importance compared to other methods in Medical image segmentation. Conventional FCM algorithm is sensitive to noise especially in the presence of intensity inhomogeneity in MRI. Main reason is that a single fuzzifier in FCM cannot properly represent pattern memberships for all clusters. In this paper, we present a novel algorithm for fuzzy segmentation of MRI data. The algorithm utilizes two fuzzifiers used in interval type-2 FCM and a spatial constraint on the membership functions. Also, in our investigation, validity functions are extended to generalized form for interval type-2 fuzzy clustering. The experimental results on both synthetic and MR images show that the proposed algorithm has better performance on image segmentation than conventional FCM based algorithms.

A new indirect approach to the type-2 fuzzy systems modeling and design

This paper proposes a new method for designing Interval Type-2 Fuzzy Logic Systems (IT2 FLSs) considering two issues: first, quality of clustering the output space and secondly, approximating the output of the IT2 FLS based on a new output processing method. Based on these two issues, we present a new cluster validity index capable of being used for type-1 Fuzzy C-Means (FCMs), Interval Type-2 FCM (IT2 FCM), and Possibilistic C-Means (PCMs) clustering algorithms. This validity index is highly efficient in determining clusters with the least similarity between them and the highest similarity between data vectors in each cluster. Then, a new definition for uncertainty bounds is presented in order to eliminate the type-reduction process in IT2 FLSs and to increase accuracy of the existing uncertainty bounds in the literature. Finally, effectiveness of the proposed approaches compared to several well-known existing methods has been investigated. Computational results have verified accuracy and effectiveness of the proposed method.

Color Textured Image Segmentation Using ICICM - Interval Type-2 Fuzzy C-Means Clustering Hybrid Approach

Segmentation is an essential process in image because of its wild application such as image analysis, medical image analysis, pattern reorganization, etc. Color and texture are most significant low-level features in an image. Normally, color-textured image segmentation consists of two steps: (i) extracting the feature and (ii) clustering the feature vector. This paper presents the hybrid approach for color texture segmentation using Haralick features extracted from the Integrated Color and Intensity Co-occurrence Matrix (ICICM). Then, Extended- Interval Type-2 Fuzzy C-means clustering algorithm is used to cluster the obtained feature vectors into several classes corresponding to the different regions of the textured image. Experimental results show that the proposed hybrid approach could obtain better cluster quality and segmentation results compared to state-of-art image segmentation algorithms.

Interval type-2 fuzzy expert system for prediction of carbon monoxide concentration in mega-cities

This paper presents an indirect approach to interval type-2 fuzzy logic system modeling to forecaste the level of air pollutants. The type-2 fuzzy logic system permits us to model the uncertainties among rules and the parameters related to data analysis. In this paper, we propose an indirect method to create an interval type-2 fuzzy logic system from a historical data, where Footprint of Uncertainties of fuzzy sets are extracted by implementation of an interval type-2 FCM algorithm and based on an upper and lower value for the level of fuzziness m in FCM. Finally, the proposed model is applied for prediction of carbon monoxide concentration in Tehran air pollution. It is shown that the proposed type-2 fuzzy logic system is superior in comparison to type-1 fuzzy logic systems in terms of two performance indices.

Interval type-2 fuzzy membership function generation methods for pattern recognition

Type-2 fuzzy sets (T2 FSs) have been shown to manage uncertainty more effectively than T1 fuzzy sets (T1 FSs) in several areas of engineering [4,6–12,15–18,21–27,30]. However, computing with T2 FSs can require undesirably large amount of computations since it involves numerous embedded T2 FSs. To reduce the complexity, interval type-2 fuzzy sets (IT2 FSs) can be used, since the secondary memberships are all equal to one [21]. In this paper, three novel interval type-2 fuzzy membership function (IT2 FMF) generation methods are proposed. The methods are based on heuristics, histograms, and interval type-2 fuzzy C-means. The performance of the methods is evaluated by applying them to back-propagation neural networks (BPNNs). Experimental results for several data sets are given to show the effectiveness of the proposed membership assignments.