Adaptive Riesz basis decomposition for image search

A novel technique via Quadtree Gauss-Laguerre Transform (LGT) for a complete localization of a complex image in an multimedia database is presented. This technique is based on an iterative Maximum Likelihood procedure that allows to compare a region of interest (ROI) of an image with a content of a database, independently from location, rotation and scale. At this aim the ROI is expanded in terms of a Riesz basis, consisting of a set Circular Harmonic Gauss Laguerre functions, selected in order to make a trade off between accuracy and computational complexity. Numerical results obtained by a Monte Carlo simulation illustrate the performance of the method.