Airborne hyperspectral imagery is valuable for military and civilian applications, such as target identification, detection
of anomalies and changes within multiple acquisitions. In target detection (TD) applications, the performance assessment
of different algorithms is an important and critical issue. In this context, the small number of public available
hyperspectral data motivated us to perform an extensive measurement campaign including various operating scenarios.
The campaign was organized by CISAM in cooperation with University of Pisa, Selex ES and CSSN-ITE, and it was
conducted in Viareggio, Italy in May, 2013. The Selex ES airborne hyperspectral sensor SIM.GA was mounted on board
of an airplane to collect images over different sites in the morning and afternoon of two subsequent days.
This paper describes the hyperspectral data collection of the trial. Four different sites were set up, representing a complex
urban scenario, two parking lots and a rural area. Targets with dimensions comparable to the sensor ground resolution
were deployed in the sites to reproduce different operating situations. An extensive ground truth documentation
completes the data collection.
Experiments to test anomalous change detection techniques were set up changing the position of the deployed targets.
Search and rescue scenarios were simulated to evaluate the performance of anomaly detection algorithms. Moreover, the
reflectance signatures of the targets were measured on the ground to perform spectral matching in varying atmospheric
and illumination conditions. The paper presents some preliminary results that show the effectiveness of hyperspectral
data exploitation for the object detection tasks of interest in this work.