Feature extraction refers to the process of screening and retrieving a set of informative and non-redundant vectors or attributes to properly characterize an observation in order to facilitate further decision analyses such as classification and pattern recognition. In general, feature extraction involves constructing distinct features to depict an object in a reduced form with sufficient accuracy, and is thus commonly considered a type of dimensionality reduction approach in image processing and pattern recognition. The development of feature extraction methods has been one of the critical issues in pattern analysis. In this chapter, an overview of feature extraction concepts and basics in association with different classification methods is presented. In addition, several accuracy metrics and indicators commonly used to assess the performance of feature extraction are also described to establish the foundation of image and data fusion.