This paper presents the results of a study designed to classify weather radar clutter echoes obtained from ground-based dual-polarization weather radar systems. The clutter signals are due to ground clutter, sea clutter, and anomalous propagation echoes, which represent sources of error in quantitative radar rainfall estimation. Fuzzy and Bayes classifiers are evaluated as an alternative approach to traditional polarimetric-based methods. Both systems were trained and validated by using C-band dual- polarization radar measurements, and a novel technique is proposed to calculate the texture function to mitigate against the edge effects at the boundaries of precipitation regions. A methodology is presented to extract the membership functions and conditional probability density functions to train the classifiers. The critical success index indicates that the Bayes classifier has, on average, a slightly better performance than the fuzzy classifiers. However, when optimal weighting was applied, the fuzzy classifier gave one of the best performances. The classifiers are sufficiently robust to be used when only single-polarization radar measurements are available.