Title | ||
---|---|---|
Improvements to Expectation-Maximization Approach for Unsupervised Classification of Remote Sensing Data |
Abstract | ||
---|---|---|
In statistical pattern recognition, mixture models allow a formal ap- proach to unsupervised learning. This work aims to present a modification of the Expectation-Maximization clustering method applied to remote sensing im- ages. The stability of its convergence has been increased by supplying the re- sults of the well-known K-Means algorithm, as seed points. Hence, the accuracy has been improved by applying cluster validity measures to each configuration, varying the initial number of clusters. High-resolution urban scenes has been tested, and we show a comparison to supervised classification results. Perfor- mance tests were also realized, showing the improvements of our proposal, in comparison to the original one. |
Year | Venue | Keywords |
---|---|---|
2007 | GeoInfo | remote sensing,mixture model,k means algorithm,unsupervised learning,high resolution,expectation maximization |
Field | DocType | Citations |
Convergence (routing),Data mining,Cluster (physics),Pattern recognition,Expectation–maximization algorithm,Computer science,Remote sensing,Unsupervised learning,Artificial intelligence,Cluster analysis,Machine learning,Mixture model | Conference | 0 |
PageRank | References | Authors |
0.34 | 4 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Thales Sehn Korting | 1 | 24 | 12.47 |
Luciano Vieira Dutra | 2 | 71 | 26.78 |
Leila Maria Garcia Fonseca | 3 | 47 | 17.89 |
Guaraci J. Erthal | 4 | 2 | 0.85 |
Felipe Castro da Silva | 5 | 5 | 0.98 |