Comments on “Principal Component Analysis of Single-Beam Echo-Sounder Signal Features for Seafloor Classification” by A. R. Amiri-Simkooei, M. Snellen, and D. G. Simons

Acoustic seabed classification with single-beam sounders is both an established process and a research topic. A recent paper (IEEE Journal of Oceanic Engineering, vol. 36, no. 2, pp. 259–271, Apr. 2011) summarizes some popular techniques, demonstrating that they can generate realistic sediment maps. In the conclusion of that paper, it is claimed that the algorithms therein can substitute for a widely used commercial software suite. This claim is not correct. Among other issues, the techniques in that paper fall short of current practice in two areas: compensation for varying water depths and choosing the best number of clusters in unsupervised classification.