The aim of this master's thesis is the classification of images of galaxies according to their morphological features using computer vision and artificial intelligence techniques. We deal specifically with the shape of the galaxy in this project. The galaxies are broadly categorized into 3 categories according to their shape: circular, elliptical and spiral. Out of these 3 possible shapes, correctly classifying the spiral shape is the most challenging. This is mostly due to the noisy images of the galaxies and partly due to the shape itself, as spiral can easily be mistaken for an ellipse or even a circle. Thus we focus on classifying the images into only 2 categories: spiral and non-spiral. The first phase of the thesis addresses the process of feature extraction from images of the galaxies, and the second phase uses artificial intelligence and machine learning methods to create a system that categorizes galaxies based on the extracted features. The specific methods used for classification are boosting, logistic regression and deep neural networks. We evaluate these techniques on data from the Galaxy Zoo project  that is freely available to anyone. The languages used are C++ (OpenCV) and Python.