Anna Tonazzini

TL;DR: In this paper, an acoustic pyrometry method for the reconstruction of temperature maps inside power plant boilers is presented based on measuring times-of-flight of acoustic waves along a number of straight paths in a cross-section of the boiler; via an integral relationship, these times depend on the temperature of the gaseous medium along the paths.

TL;DR: A new model for bleed-through in grayscale document images is proposed, based on the availability of the recto and verso pages, and it is shown that blind source separation can be successfully applied in this case too.

TL;DR: A novel approach to restoring digital document images, viewing the problem as one of separating overlapped texts and then reformulating it as a blind source separation problem, approached through independent component analysis techniques, which have the advantage that no models are required for the background.

TL;DR: In this paper, an independent component analysis (ICA) algorithm is proposed to separate signals of different origin in sky maps at several frequencies. But it works without prior assumptions on either the frequency dependence or the angular power spectrum of the various signals; rather, it learns directly from the input data how to identify the statistically independent components, on the assumption that all but one of the components have non-Gaussian distributions.

TL;DR: This paper proposes an expectation-maximization algorithm with the mean field approximation to derive a procedure for estimating the mixing matrix, the sources, and their edge maps, and finds that a source model accounting for local autocorrelation is able to increase robustness against noise, even space variant.