Optical Coherence Tomography (OCT) is a fiber--optic imaging modality which produces high resolution tomographic images of the coronary lumen and outer vessel wall. While OCT images present morphological information in highly resolved detail, the characterization of the various plaque components relies on trained readers. The aim of this study is to extract a set of features in grayscale OCT images and to use them in order to classify the atherosclerotic plaque. Intensity and texture based features we used in order to classify the plaque in four plaque types: Calcium (C), Lipid Pool (LP), Fibrous Tissue (FT) and Mixed Plaque (MP). 50 OCT annotated images from 3 patients were used to train and test the proposed plaque characterization method. Using a Random Forests classifier overall classification accuracy 80.41% is reported.