Extraction of Road Geometric Parameters from High Resolution Remote Sensing Images Validated by GNSS/INS Geodetic Techniques

Roadway geometric data, user behavior and crash data provide the main input for developing existing highway safety evaluations. In particular, roadway geometric data can be useful in providing a quantitative guidance for alignment consistency and for having an initial indication of critical point pres-ence along a road. Furthermore the article 13 of the Italian Road Code requires an implementation of Road Cadastre for which it is necessary to know the planimetric and altimet-ric alignments. This paper presents a methodology to obtain geometric road information from high resolution images using automatic techniques for the extraction of road geometric parameters. The paper starts from the Road Cadastre characteristics and definitions and the analysis of the existing studies regarding remote sensing methodologies. Then two strategies, tested on road sections in Italy, are presented. The first one is based on the concept of spectral signature, while the second one is based on the Fractal Dimension of the image treated only as a numerical matrix. The results were validated by GNSS/INS geodetic techniques, using also real time kinematic data.