Vehicle Make and Model classification system using bag of SIFT features

摘要

Vehicle Make and Model classification is an important part of Intelligent Transportation System. In this work, we have proposed a method based on Linear Support vector machine to solve this problem. Scale Invariant Transform Feature (SIFT) algorithm is used in this work to extract and represent local interest points. Bag of words model is used to represent the local features as fixed length vector to represent an image. The proposed method is evaluated on a publicly available vehicle make and model dataset and promising results are achieved with this method.