Linear Feature Extraction from High-Resolution SAR Images

In this book, linear feature extraction issue is firstly tackled by detecting amplitude discontinuities (i.e. edges) of the signal. Next, the obtained result is refined by an edge linking stage so that, finally, the boundary of the object of interest is extracted throughout a higher-level model. Initially, the speckle model is exploited for improving edge detection operation. Next, to free the process by statistical assumptions of data, general multiscale linear filtering is investigated. The edge linking stage makes use of the sequential edge linking (SEL) algorithm with two novel metrics useful to cope with the one-look SAR data. The algorithm devised to extract linear features (e.g. roads and runways), relies on the Hough Transform framework and tries reconstructing the object boundaries as composition of linear segments. Finally, to improve edge detection performance and approach a higher level of processing, a novel despeckling algorithm is presented. It belongs to the class of non-linear, anisotropic diffusion filters that apply a partial differential equation (PDE) onto the image in analysis.

Autorentext

Luca Fabbrini received the Eng. degree in the 2010 with 110/110 cum laude from the University of Florence, Italy. After receiving the Ph.D. in Electronic System Engineering from the same university he joined the R&D radar laboratory at IDS S.p.A. fulfilling the position of signal and image processing analyst.