Based on the extraction of texture features, the Bayesian decision rule is employed to identify the decision threshold that separates the target from the background in the magnitude image. Then, the training samples for the SVDD classifier are automatically selected and used to train the classifier. Finally, the trained SVDD classifier is used to classify the rest pixels of the thresholding process. Experimental results obtained on real and simulated SAR imageries demonstrate the effectiveness of the proposed method. © Springer-Verlag Berlin Heidelberg 2011.
CITATION STYLE
Zhang, X., Song, J., Yi, Z., & Wang, R. (2011). Unsupervised SAR imagery segmentation based on SVDD. In Advances in Intelligent and Soft Computing (Vol. 128, pp. 25–31). https://doi.org/10.1007/978-3-642-25989-0_5
Mendeley helps you to discover research relevant for your work.