Improving and simulating urban landscape image recognition using combination optimization and fuzzy K-means algorithm

This paper presents the development of generalized fuzzy k-means algorithms and their application in image compression based on vector quantization. The development of generalized fuzzy k-means algorithms is based on the search for partitions of the feature vector space other than those generated by existing fuzzy k-means algorithms. These alternative partitions can be obtained by relaxing one of the conditions imposed on the membership functions. The clustering problem is formulated as a constrained minimization problem, whose solution depends on the selection of a constrain function that satisfies certain conditions. The solution of this minimization problem results in a broad family of generalized fuzzy k-means algorithms, which include the existing fuzzy k-means algorithms as a special case. Moreover, the proposed formulation results in the minimum fuzzy k-means algorithms, which are computationally less demanding than the existing fuzzy k-means algorithms. A broad family of admissible constrain functions result in an extended family of fuzzy k-means algorithms, which at the limit provide the fuzzy k-means and minimum fuzzy k-means algorithms. The resulting algorithms are used in image compression based on vector quantization.

In this paper, we proposed two hybrid data clustering algorithms that are called ICAFKM and PSOFKM. ICAFKM combined the advantageous aspects of Fuzzy K-Means (FKM) and Imperialist Competitive Algorithm (ICA), and PSOFKM makes full use of the merits of both Particle Swarm Optimization (PSO) and FKM algorithms. FKM is one of the most popular data clustering methods. However, this algorithm solves the problem of sensitivity to initial states of K-Means (KM) algorithm, but like KM, it often converges to local optima. The proposed ICAFKM and PSOFKM algorithms aim to help the FKM to escape from local optima and increase the convergence speed of the ICA and PSO algorithms in clustering process. In order to evaluate the performance of ICAFKM and PSOFKM methods, we evaluate these algorithms on five datasets and compared them with FKM, ICA, PSO, PSOKHM, and HABC algorithms. The experimental results indicate that the ICAFKM carries out better results than the other methods.

Fuzzy k-means clustering algorithms have successfully been applied to digital image segmentations and classifications as an improvement of the conventional k-means cluster algorithm. The limitation of the Fuzzy k-means algorithm is its large computation cost. In this paper, we propose a Successive Over-Relaxation (SOR) based fuzzy k-means algorithm in order to accelerate the convergence of the algorithm. The SOR is a variant of the Gauss–Seidel method for solving a linear system of equations, resulting in faster convergence. The proposed method has been applied to classification of remotely sensed images. Experimental results show that the proposed SOR based fuzzy k-means algorithm can improve convergence speed significantly and yields comparable similar classification results with conventional fuzzy k-means algorithm.

This paper presents a comparison study of the fuzzy k-means algorithm and a new variant with variable weighting in clustering high dimensional data. The fuzzy k-means algorithm is effective in discovering the clusters with overlapping boundaries. However, this effectiveness can be handicapped in high dimensional data. The recent development of the k-means algorithm with automated variable weighting offers a new technique for dealing with high dimensional data that occurs in many new applications such as text mining and bioinformatics. In this paper, the variable weighting mechanism is incorporated in the fuzzy k-means algorithm to cluster high dimensional data with overlapping clusters. Experiments on real data sets have shown that the variable weighting fuzzy k-means produced better clustering results than the fuzzy k-means without variable weighting.

One of the main drawbacks of the fuzzy K-means algorithm, when applied on color image segmentation is its low speed, which is due to its exhaustive processing mechanism. In this paper we present a methodology for performing fast color clustering using the fuzzy K-means algorithm. Instead of applying the fuzzy K-means algorithm on the entire image we do that on a small sample subset of the initial image. Two alternative sampling approaches are applied and compared, namely, the random sampling and a sampling based on Hilbert's algorithm. It is experimentally shown that, when using only the sampled subset of the image, the speed gain is very high. The classification results, at the same time, are quite similar to those obtained by using the entire image, with the second sampling approach presenting much better results.

Although fuzzy k-modes algorithm has removed the numeric-only limitation of the k-means algorithm, that each attribute of the centroid with a single category value and the use of a simple distance measure will compromise its precision, and therefore prone to falling into local optima. In this paper, an extended fuzzy k-means(xFKM) algorithm for clustering categorical valued data is presented, in which the cluster centroid vectors are represented as expanded forms to keep more clustering information as possible as well, and updated with the method similar to fuzzy k-means algorithm. Experiments on several real databases show that xFKM algorithm can get better clustering result than fuzzy k-modes algorithm does.

The k-AMH algorithm has been proven efficient in clustering categorical datasets. It can also be used to cluster numerical values with minimum modification to the original algorithm. In this paper, we present two algorithms that extend the k-AMH algorithm to the clustering of numerical values. The original k-AMH algorithm for categorical values uses a simple matching dissimilarity measure, but for numerical values it uses Euclidean distance. The first extension to the k-AMH algorithm, denoted k-AMH Numeric I, enables it to cluster numerical values in a fashion similar to k-AMH for categorical data. The second extension, k-AMH Numeric II, adopts the cost function of the fuzzy k-Means algorithm together with Euclidean distance, and has demonstrated performance similar to that of k-AMH Numeric I. The clustering performance of the two algorithms was evaluated on six real-world datasets against a benchmark algorithm, the fuzzy k-Means algorithm. The results obtained indicate that the two algorithms are as efficient as the fuzzy k-Means algorithm when clustering numerical values. Further, on an ANOVA test, k-AMH Numeric I obtained the highest accuracy score of 0.69 for the six datasets combined with p-value less than 0.01, indicating a 95% confidence level. The experimental results prove that the k-AMH Numeric I and k-AMH Numeric II algorithms can be effectively used for numerical clustering. The significance of this study lies in that the k-AMH numeric algorithms have been demonstrated as potential solutions for clustering numerical objects.

A fuzzy K-Means algorithm based on Fisher distribution for the identification of rock discontinuity sets

In this paper, we propose a new data clustering method using local probability and hypothesis theory. To cluster the test data set we analyze the local area of the test data set using local probability distribution and decide the candidate class of the data set using mean standard deviation and variance etc. To decide each class of the test data, statistical hypothesis theory is applied to the decided candidate class of the test data set. For evaluating, the proposed classification method is compared to the conventional fuzzy c-mean method, k-means algorithm and Discriminator analysis algorithm. The simulation results show more accuracy than results of fuzzy c-mean method, k-means algorithm and Discriminator analysis algorithm.

This paper proposes a distributed wireless sensor network (WSN) data stream clustering algorithm to minimize sensor nodes energy consumption and consequently extend the network lifetime. The paper follows the strategy of trading-off communication for computation through distributed clustering and successive transmission of local clusters. We present an energy efficient algorithm we developed, subtractive fuzzy cluster means (SUBFCM), and analyze its energy efficiency as well as clustering performance in comparison with state-of-the-art standard data clustering algorithms such as fuzzy c-means and k-means algorithms. Simulations show that SUBFCM can achieve WSN data stream clustering with significantly less energy than that required by fuzzy c-means and k-means algorithms.

In this paper, we propose a new classification method using local probability and statistical hypothesis theory. To separate the test data, we analyze the local area of the test data set using local probability distribution and decide the candidate class of the data set. To decide each class of the test data, statistical hypothesis theory is applied to the decided candidate class of the test data set. To evaluate, the proposed classification method is compared to the conventional fuzzy c-mean method and k-means algorithm. The simulation results show more accuracy than results of fuzzy c-mean method and k-means algorithm.

A Unified Form of Fuzzy C-Means and K-Means algorithms and its Partitional Implementation

ABSTRACTDelineation of site-specific nutrient management zones (MZ) provides a basis for practical and cost-effective management of spatial soil fertility in precision agriculture. Therefore, the objective of this study was the delineation of MZs in a soybean field using geostatistics, principal component analysis (PCA), and the fuzzy k-means algorithm. The study was carried out in a field with 204 ha located in São Desidério, western Bahia state, Brazil (12 ° 25 ‘ 12” S, 45 ° 29ʹ 46” W). To do so, samples of soil attributes (0–20 cm), soybean yield, electrical conductivity (EC) at 0.20 m (EC02), 0.50 m (EC05), 1.00 m (EC1), 2.00 m (EC2) soil profile depth, and the Normalized Difference Vegetation Index (NDVI) were obtained in 204 points (100 x 100 m grid). After soil sampling and laboratory analyzes, the data were submitted to descriptive statistics and a Spearman correlation analysis was performed to select those attributes related to soybean yield. Then, the spatial variability of these attributes was assessed and spatial distribution maps were constructed using geostatistical tools. Next, PCA and fuzzy k-means algorithm were then performed to delineate MZs. Finally, the agreement between the MZs maps obtained from the PCA and soybean yield was assessed using the Kappa index. Results showed that the optimal number of MZs was two, which resulted in a Kappa index of 0.61 (very good). Moreover, the analysis of variance indicated heterogeneity between all attributes analyzed in the MZs. Finally, the defined MZs provide a basis of information for site-specific nutrient management.

The method of latent semantic indexing (LSI) is an information retrieval technique using a low-rank singular value decomposition (SVD) of the term-document matrix. Although the LSI method has empirical success, it suffers from the lack of interpretation for the low-rank approximation and, consequently, the lack of controls for accomplishing specific tasks in information retrieval. A method introduced by Dhillon and Modha is an improvement in that direction. It uses centroids of clusters or so called concept decomposition for lowering the rank of the term-document matrix. We focus on improvements of that method using fuzzy k-means algorithm. Also, we compare the precision of information retrieval for the two above methods.

<p>In this article, a combined optimization algorithm was proposed which combines the optimal adaptive Cuckoo algorithm (OACS) which is Nature-inspired algorithm with Gray Wolf optimizer algorithm (GWO). Sometimes considering the cuckoo algorithm alone, may fail to find the local minimum-point and also fails to reach to the solution because of the slow speed of its convergence property. Therefore, considering the new proposed adaptive combined algorithm gave a strong improvement for using this to reach the minimum point in solving (23) nonlinear test problems. This is suitable to solve a large number of nonlinear unconstraint optimization test functions with obtaining good and robust numerical results.</p>