Search

Soft classification using Kohonen's Self-Organizing Map (SOM) has not been explored as thoroughly as the Multi-Layer-Perceptron (MLP) neural network. In this paper, we propose two non-parametric algorithms for the SOM to provide soft classification outputs. These algorithms, which are labelling-frequency-based, are called SOM Commitment (SOM-C) and SOM Typicality (SOM-T), expressing in the first case the degree of commitment the classifier has for each class for a specific pixel and in the second case, how typical that pixel's reflectances are of those upon which the classifier was trained. To evaluate the two proposed algorithms, soft classifications of a Satellite Pour l'Observation de la Terre (SPOT) High Resolution Visible (HRV) image and an Airborne Visible Infrared Imaging Spectrometer (AVIRIS) image were undertaken. Both traditional soft classifiers, i.e. Bayesian posterior probability and Mahalanobis typicality classifier, and the most frequently used non-parametric neural network model, i.e. MLP, were used as a comparison. Principal-components analysis (PCA) was used to explore the relationship between these measures. Results indicate that great similarities exist between the SOM-C, MLP and the Bayesian posterior probability classifiers, while the SOM-T corresponds closely with Mahalanobis typicality probabilities. However, as implemented, they have the advantage of being non-parametric. The proposed measures significantly outperformed Bayesian and Mahalanobis classifiers when using the hyperspectral AVIRIS image. [ABSTRACT FROM AUTHOR]

In this paper, a maximum likelihood (ML) classification algorithm is proposed to classify multi-look polarimetric SAR images. This algorithm considers a generalized multiplicative speckle model in which three texture variables are assumed to affect separately three polarization channels. We derive the ML estimation of the texture parameters for each polarization channel with the conditional complex Wishart distribution of the multi-look polarimetric covariance matrix, and design the corresponding M L classifier according to the Bayesian criterion. Further, a method for adaptively producing the a priori probabilities is presented in order to improve the classification accuracy. This method utilizes the contextual information in a forward procedure, and does not need any iteration. With the NASA/JPL L band four-look polarimetric SAR data, the effectiveness of the classification algorithm is demonstrated, and the use of the adaptive a priori probabilities is shown to result in improved classifications. [ABSTRACT FROM AUTHOR]