Multi-scale classification for electro-sensing

13.01.2021 15:00 - 16:00

Andrea Scapin

Abstract: The biological behavior of weakly electric fish has been studied by scholars for years. These fish orient themselves at night in complete darkness by using electrosensory information, which makes these animals an ideal subject for developing bio-inspired imaging techniques. Such interest has motivated a huge number of studies addressing the active electro-sensing problem from many different perspectives since Lissmann and Machin’s work. One of the most noteworthy potential bio-inspired applications is in underwater robotics. Building autonomous robots with electro-sensing technology may supply unexplored navigation, imaging and classification capabilities, especially when the sight is unreliable due, for example, to the turbidity of the surrounding waters or the poor lighting conditions.

In this talk we present a premier and innovative (real-time) multi-scale method for target classification in electro-sensing. The intent is that of mimicking the behavior of the weakly electric fish, which is able to retrieve much more information about the target by approaching it. The method is based on a family of transform-invariant shape descriptors computed from generalized polarization tensors (GPTs) reconstructed at multiple scales. The evidence provided by the different descriptors at each scale is fused using Dempster-Shafer Theory.

Online unter

Online via BigBlueButton