Goal of this work is to present and validate an underwater object spectral recognition methodology for fluorescence LIDAR signals by using an underwater fluorescence LIDAR propagation model. The spectral recognition methodology is aimed at deciding if an underwater object detected in the water column can be identified as belonging to a data base of objects of interest characterized by known fluorescence spectral signatures. The methodology needs to compensate the received signal for the water column effects in order to derive an estimate of the underwater object fluorescence spectrum to be used for spectral recognition. By using an underwater fluorescence LIDAR propagation model developed ad hoc, the methodology may be validated in different system, geometric, and environmental conditions. Experimental results obtained in two different acquisition scenarios show that the underwater object recognition methodology is promising for recognizing objects submerged in the water column at different depths and highlight the utility of the developed LIDAR propagation model for assessing the object recognition performance that may be experienced in various different acquisition conditions.
Fluorescence LIght Detection And Ranging (LIDAR) systems have been proven powerful for detecting and recognizing underwater objects in several applications. Such Fluorescence systems have been employed mainly for detecting and recognizing oil spill and chemicals dissolved in the sea and to identify phytoplankton species. This work focuses on the use of Fluorescence LIDAR systems in underwater object recognition applications. In fact, the fluorescence spectra induced over object and materials may be exploited to derive chemical-physical information about object nature useful to recognition. Specifically, a model for fluorescence LIDAR transmission in the water medium, both in the presence and absence, of an underwater object is proposed. The developed model describes the interaction of the transmitted laser beam with underwater objects, bottom, and water molecules. Specifically, the fluorescence return signals are modeled involving the inelastic backscattering contributions due to the Raman scattering by water molecules and fluorescence by water constituents, bottom, and objects. A range of simulations have been performed modeling the immersion of an object at different depths within the water column for a variety of system characteristics and water environmental conditions. Simulation results show the model flexibility for reproducing the signals acquired in different operational scenarios on the basis of various system parameters, acquisition geometries, and water environments. The transmission model may be useful to predict the performance of a given fluorescence LIDAR in specific underwater object detection and recognition applications.