D Latini

RFIDrone: Preliminary experiments and electromagnetic models

We introduce in a unitary way the paradigm of radiofrequency identification (RFID) merged with the technology of Unmanned Aerial Vehicles (UAV) giving rise to RFIDrone devices. Such family comprises the READER-Drone, which is a suitable UAV integrated with an autonomous RFID reader to act as mobile scanner of the environment, and the TAG-Drone, a UAV only equipped with an RFID sensor tag that hence becomes a mobile and automatically re-positioned sensor. We shows some handy electromagnetic models to identify the upper-bound communication performance of RFIDrone in close proximity of a scattering surface and we resume the results of some preliminary open-air experimentation corroborating the theoretical analysis.

Oil spill analysis by means of full polarimetric UAVSAR (L-band) and Radarsat-2 (C-band) products acquired during Deepwater Horizon Disaster

SAR instruments with polarimetric capabilities, high resolution and short revisit time can provide powerful support in oil spill monitoring and different techniques of analysis have been developed for this purpose [1][2]. An oil film on the sea surface results in darker areas in SAR images, but careful interpretation is required because dark spots can also be caused by natural phenomena. In view of the very low backscatter from slicks, the Noise Equivalent Sigma Zero (NESZ) is a primary sensor parameter to be considered when using a sensor for slick analysis. Among the existing full polarimetric sensors, the high resolution and very low NESZ values of UAVSAR (L-band) and RADARSAT-2 (C-band) make them preferable for oil spill analysis compared to the last generation SAR instruments. The Deepwater Horizon disaster that occurred in the Gulf of Mexico in 2010 represents a unique and extensive test site where large amounts of SAR imagery and ground validation data are available. By applying the Cloude-Pottier decomposition method to full polarimetric UAVSAR (L-band) and RADARSAT-2 (C-band), it is possible to extract parameters that describe the scattering mechanism of the target. By comparing quasi-simultaneous acquisitions and exploiting the different penetration capabilities of the sensors, we investigate the potential of full polarimetric SAR to discriminate oil on the sea surface from look-alike phenomena covering the full range of backscattering values down to those at the instrument noise floor.

On neural networks algorithms for oil spill detection when applied to C- and X-band SAR

The aim of this paper is to introduce new algorithms for the oil spill detection taking fully advantage of the polarimetric and textural features contained in new generation SAR data such as those provided by Radarsat-2 and COSMO-SkyMed missions. The SAR information is exploited using a new statistical decomposition method based on AANN. Thanks to the AANN the original image is represented in terms of Nonlinear principal components (NLPC). The oil spill detection procedure is then directly applied to the new generated components.