Development of a fusion adaptive algorithm for marine debris detection and classification within the post-Sandy restoration framework
|Title||Development of a fusion adaptive algorithm for marine debris detection and classification within the post-Sandy restoration framework|
|Publication Type||Conference Abstract|
|Authors||Masetti, Giuseppe, and Calder, Brian R.|
|Conference Name||Canadian Hydrographic Conference 2014|
|Conference Dates||15-17 April 2014|
|Conference Location||St. John's, Canada|
|Keywords||emergency response, fusion adaptive algorithm, marine debris, target classification, target detection|
Identification and classification of marine debris represent a difficult task due to the extreme variability of the marine environment, the possible targets, and the variable skill levels of human operators. In particular, the range of potential targets is much wider than similar fields of research such as mine hunting, localization of unexploded ordnance or pipeline detection. In order to address this additional complexity, an adaptive algorithm is being developing that appropriately responds to changes in the environment, context, and human skills.
The preliminary step is to properly geometrically and radiometrically correct the collected data. Then, the core engine manages the fusion of a set of statistically- and physically-based algorithms, working at different levels (swath, beam, snippet, and pixel) and using both predictive modeling and phenomenological approaches. The main outcome is the reduction of inter-algorithmic cross-correlation and, thus, the probability of false alarm. In particular, this characteristic is particularly relevant in the case of areas covered by single-device data sets. At this early stage, we provide a proof of concept showing outcomes from an algorithm that dynamically adapts itself to the depth and average backscatter level met in the surveyed environment, targeting marine debris that are modeled as objects of about 1-m size.
The project relies on a modular software library, called MaTaDOR (Marine Target Detection and Object Recognition). The initially supported data formats are Kongsberg EM series and SAIC Generic Sensor Format (GSF), but the plan is to enlarge the range to the more common formats (taking advantage of the parallel ongoing HUDDL project).