Galaxies in the universe are commonly classified by their morphology, or visual appearance. The morphology of a galaxy tells us about the history and physical make-up of the galaxy. With the fast pace at which digital galaxy images are captured and a slow and biased human pattern recognition process, finding an efficient way to automate the galaxy image classification process can help advance the knowledge toward understanding the universe. In this paper, we describe a machine learning approach toward galaxy image classification. We use an ensemble of Support Vector Machines to classify galaxy images found in the Sloan Digital Sky Survey into one or more of the thirty-seven morphological categories used in the Galaxy Zoo 2 project. The preliminary results of our approach compare favorably to those of previous work.