An unsupervised method of classifying remotely sensed images using Kohonen self‐organizing maps and agglomerative hierarchical clustering methods

Abstract

Unlike conventional unsupervised classification methods, such as K-means and ISODATA, which are based on partitional clustering techniques, the methodology proposed in this work attempts to take advantage of the properties of Kohonen's self-organizing map (SOM) together with agglomerative hierarchical clustering methods to perform the automatic classification of remotely sensed images. The key point of the proposed method is to execute the cluster analysis process by means of a set of SOM prototypes, instead of working directly with the original patterns of the image. This strategy significantly reduces the complexity of the data analysis, making it possible to use techniques that have not normally been considered viable in the processing of remotely sensed images, such as hierarchical clustering methods and cluster validation indices. Through the use of the SOM, the proposed method maps the original patterns of the image to a two-dimensional neural grid, attempting to preserve the probability distribution and topology of the input space. Afterwards, an agglomerative hierarchical clustering method with restricted connectivity is applied to the trained neural grid, generating a simplified dendrogram for the image data. Utilizing SOM statistic properties, the method employs modified versions of cluster validation indices to automatically determine the ideal number of clusters for the image. The experimental results show examples of the application of the proposed methodology and compare its performance to the K-means algorithm.

Support vector machines in remote sensing: A review

“…Of particular interest would be comparison/fusion with algorithms such as self-organizing maps (Kohonen, 1997) that address efficiently high dimensionality problems and have already found fruitful ground in remote sensing (e.g., Hong et al, 2006;Goncalves et al, 2008). In addition, integration with methodologies that deal more naturally with multi-class problems without the SVM complexity may further advance SVM understanding, for example a learning vector quantization system (Schneider et al, 2009).…”

Remote Sensing Exploration of Nb-Ta-LREE-Enriched Carbonatite (Epembe/Namibia)

“…The data-driven algorithm belongs to the group of unsupervised artificial neural networks and can handle non-linear relationships and non-Gaussian data distributions [47]. SOMs get trained by competitive learning and are perfectly suited for hyperspectral data analysis, as they outperform most unsupervised algorithms ( [3,48,49]). …”

PartSOM: A Framework for Distributed Data Clustering Using SOM and K-Means

“…Each resulting neuron cluster is a sub-graph that defines complex and non-parametric geometries in the input space, which approximately describes the shape of the clusters. Regarding the last approach, Gonçalves et al (2008) present improvements of contiguity-constrained hierarchical clustering approaches using validation indexes. Some works apply clustering algorithms over the U-matrix to segment the map in well defined regions.…”

“…A host of strategies for cluster detection using the U-matrix were proposed in the literature (Costa & Andrade Netto, 2001a;Costa & Andrade Netto, 2003). Three main algorithms were presented: mathematical morphology derived map segmentation (Costa & Andrade Netto, 2001b); a graph partitioning approach (Costa & Andrade Netto, 2003) and contiguityconstrained hierarchical clustering approaches (Gonçalves et al, 2008;Murtagh, 1995). Both algorithms were developed for automatic partitioning and labeling of a trained SOM network.…”