Visual indexing and retrieval

The second chapter provides a deep overview of the basic visual feature extraction and description methods and approaches in the literature for images as well as for spatio-temporal data analysis. The chapter aims to clarify the process of the main recognition steps of detection and description of image. Interest point detection approaches, point-based and region-based detectors, the spatiotemporal feature extraction and some reference feature descriptors are also presented and discussed. On the rest, some well-known feature extraction schemes, like Harris and LoG detectors and feature description approaches like SIFT, SURF and GIST are presented.