Image Classification with Local Features

This paper presents a novel method for classify images using local features in cluttered real-world scenes. After analyzing of several fashionable local features at present, we choose the suitable features to construct the visual vocabulary. These visual words are invariant to image scale and rotation, and are shown robust to addition of noise and changes in 3D viewpoint. We also describe two approaches to represent images using these visual words. As baselines for comparison, some additional classification systems also have been implemented. The performance analysis on the obtained experimental results demonstrates that the proposed methods are effective and efficient.