A web-based environment for shape analysis of fish otoliths. The AFORO database

Abstract

We present an automatic system for describing and identifying fish otolith shapes combining information technologies with modern advances in signal analysis. As a major novelty with respect to past studies the system includes wavelets (WT) and curvature space scale (CSS) analyses that have been shown to provide local morphologic information and characteristics (landmarks) of the sagitta otolith outline for non-specialists. By applying this electronic catalogue directly the system is very helpful for determining teleostean fish size, population and species. Both the electronic catalogue and the shape analysis are implemented through the AFORO database, which at present contains more than 1250 high resolution images of the left sagitta from 292 species and 94 families of teleostan fishes sampled mainly in Mediterranean and Antarctic waters.

Keywords

• otolith
• shape analysis
• curvature scale space
• wavelets
• fourier