Multiobjective Genetic Clustering for Pixel Classification in Remote Sensing Imagery

Abstract

An important approach for unsupervised landcover classification in remote sensing images is the clustering of pixels in the spectral domain into several fuzzy partitions. In this paper, a multiobjective optimization algorithm is utilized to tackle the problem of fuzzy partitioning where a number of fuzzy cluster validity indexes are simultaneously optimized. The resultant set of near-Pareto-optimal solutions contains a number of nondominated solutions, which the user can judge relatively and pick up the most promising one according to the problem requirements. Real-coded encoding of the cluster centers is used for this purpose.Results demonstrating the effectiveness of the proposed technique are provided for numeric remote sensing data described in terms of feature vectors. Different landcover regions in remote sensing imagery have also been classified using the proposed technique to establish its efficiency.Index Terms-Cluster validity measures, fuzzy clustering, genetic algorithm (GA), multiobjective optimization (MOO), Pareto-optimal, pixel classification, remote sensing imagery.

Oil in Syria between Terrorism and Dictatorship

“…While selection is random any individual has the choice to become a parent, selection is clearly biased towards fitted individuals. 64 Parents are not required to be distinctive for any iteration; fit individuals may produce many offspring's the crossover is selected at random and mutation is applied to all individuals in the new population. With probability P m , each bit on every string is inverted.…”

“…69 2013 MOPs are used to embark upon the arduousness of fuzzy partitioning where a number of fuzzy cluster validity indices are simultaneously optimized. The resultant set of near-Pareto-optimal answers contains a number of non-dominated answers, which the user can referee relatively and choose upon the most undertaking one according to the problem necessity; Real-coded encoding of the cluster centers is utilized for this principle [64].…”

Interactive Classification of Remote Sensing Images by Using Optimum-Path Forest and Genetic Programming

“…Apart from the classification methods presented in [14,12,3], several others have been proposed recently [8,13,1,10,2]. The novelty of these approaches relies on: resolution and number of bands of ISRs; type of extracted features; used learning technique, and level of discrimination among the classes of the image (some studies include all the vegetation types in the same class, for example).…”

“…al [13] and Bandyopadhyay et. al [1] use Genetic algorithms (GA). The former uses GA to find the configuration parameters of a neural network while the latter uses GA for clustering the pixels of the RSIs.…”

“…1 Compute the distance d(s, t) from the descriptors for every subimage t ∈Î. 2 Create an ordered list L of the N closest subimages t to s based on d(s, t).…”