In this paper, we propose a prototype classification method that employs a learning process to determine both the number and the location of prototypes. This learning process decides whether to stop adding prototypes ...
详细信息
In this paper, we propose a prototype classification method that employs a learning process to determine both the number and the location of prototypes. This learning process decides whether to stop adding prototypes according to a certain termination condition, and also adjusts the location of prototypes using either the K-means (KM) or the fuzzyc-means (FcM) clusteringalgorithms. When the prototype classification rnethod is applied, the support vector machine (SVM) method can be used to post-process the top-rank candidates obtained during the prototype learning or matching process. We apply this hybrid solution to handwriting recognition and address the convergence behavior and runtime consumption of the prototype construction process, and discuss how to combine our prototype classifier with SVM classifiers to form an effective hybrid classifier. (c) 2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
This paper presents a method of fuzzyclustering based on niche geneticalgorithm in order to solve the clustering problem of grid nodes resource. The method inducts niche geneticalgorithm into fuzzyclustering by us...
详细信息
This paper presents a method of fuzzyclustering based on niche geneticalgorithm in order to solve the clustering problem of grid nodes resource. The method inducts niche geneticalgorithm into fuzzyclustering by using its searches randomly and parallelism, which instructs to choose the number of cluster centers and data that are cluster centers. It resolves the problem on sensitiveness of the initial condition of fuzzyc- meansclustering. Experiment results show that the method has global convergence, avoids local minimum value, sorts effectively grid nodes resource and improve performance of grid resource discovery.
Spectroscopic image analysis can be characterized as two different and distinct problems depending on the kind of information required from the solution. In its simplest form, the data can be decomposed into two subma...
详细信息
Spectroscopic image analysis can be characterized as two different and distinct problems depending on the kind of information required from the solution. In its simplest form, the data can be decomposed into two submatrices, each of which carries different aspects of pure component information. Typically, this means information about pure spectra and pure intensities are obtained from the solution. This is the well-known and well-characterized bilinear form that by itself cannot guarantee a unique solution due to the rotational ambiguity inherent in the mathematical solution. This problem has been addressed in a number of ways by different authors using novel constraints applied to the least-squares solution. This usually takes the form of natural constraints as suggested by Tauler as the standard methodology to improve the resolution of data during alternative least squares (ALS) iterative process. A second type of multivariate image analysis problem is proposed in this paper that is quite different from the tradition methods and in some ways potentially is more useful. This involves the solution as a class problem in which the relevant information is not necessarily contained in pure component information, but rather, in unique combinations of the pure components that are allowed to be spatially collocated. This discriminant image resolution (DIR) method theoretically can be treated as a more generalized solution to the problem because the distribution of components is allowed to freely mix in simplified combinations of solutions. The result is a constrained least-squares solution where the constraints are more limited and therefore less restrictive. The constraints in this case employ the results of probabilisticclass partition information by applying Bayesian discriminant clustering to the intensity submatrix. This amounts to a spatial constraint because the probability of class association is used as a way of limiting the components that are allowed to appear
An index model for quality evaluation based on the formula of similarity membership functions in the fuzzyc-means (FcM) clusteringalgorithm is proposed. Summing up the weighted similarity degrees between an observat...
详细信息
An index model for quality evaluation based on the formula of similarity membership functions in the fuzzyc-means (FcM) clusteringalgorithm is proposed. Summing up the weighted similarity degrees between an observation and designed specific quality-ordered levels develops an alternative overall index. Stretching the values of the controlling parameters in the formula of the similarity membership functions causes diverse patterns of overall index models. Applying this proposed fuzzy index model to the trophic evaluation of reservoir waters is studied to demonstrate the practical application of this index. Every measurement of the variables is standardized by the membership function of quality evaluation on the interval [0,1], referring to the trophic status clarified in the carlson Trophic State Index. The sensitivity analyses are studied both in the proposed index system and the carlson Trophic State Index. Besides, a case study of the trophic status evaluation of the Feitsui Reservoir from 1987 to 2003 is presented to demonstrate the feasibility of applying the proposed evaluation model. (c) 2005 Elsevier Ltd. All rights reserved.
fuzzy c-means clustering algorithm is a classical non-supervised classification *** image classification, fuzzy c-means clustering algorithm makes decisions on a pixel-by-pixel basis and does not take advantage of spa...
详细信息
fuzzy c-means clustering algorithm is a classical non-supervised classification *** image classification, fuzzy c-means clustering algorithm makes decisions on a pixel-by-pixel basis and does not take advantage of spatial information, regardless of the pixels' correlation. In this letter, a novel fuzzy c-means clustering algorithm is introduced, which is based on image's neighborhood system. During classification procedure, the novel algorithm regards all pixels'fuzzy membership as a random field. The neighboring pixels' fuzzy membership information is used for the algorithm's iteration procedure. As a result, the algorithm gives a more smooth classification result and cuts down the computation time.
Based on the fuzzyclustering method, we improved a neuro-fuzzy learning algorithm. In this improved approach, before learning fuzzy rules we extract typical data from training data by using fuzzyc-meansclustering a...
详细信息
Based on the fuzzyclustering method, we improved a neuro-fuzzy learning algorithm. In this improved approach, before learning fuzzy rules we extract typical data from training data by using fuzzy c-means clustering algorithm, in order to remove redundant data and resolve conflicts in data, and make them as practical training data. By these typical data, fuzzy rules can be tuned by using the neuro-fuzzy learning algorithm. Therefore, the learning time can be expected to be reduced and the fuzzy rules generated by the improved approach are reasonable and suitable for the identified system model, Moreover, the efficiency of the improved method is also shown by identifying nonlinear functions. (c) 2001 Elsevier Science B.V. All rights reserved.
This paper presents a fuzzyclusteringalgorithm for the extraction of a smooth curve from unordered noisy data. In this method, the input data are first clustered into different regions using the fuzzyc-means algori...
详细信息
This paper presents a fuzzyclusteringalgorithm for the extraction of a smooth curve from unordered noisy data. In this method, the input data are first clustered into different regions using the fuzzyc-meansalgorithm and each region is represented by its cluster center. Neighboring cluster centers are linked to produce a graph according to the average class membership values. Loops in the graph are removed to form a curve according to spatial relations of the cluster centers. The input samples are then reclustered using the fuzzyc-means (FcM) algorithm, with the constraint that the curve must be smooth. The method has been tested with both open and closed curves with good results.
暂无评论