Fully-automated identification of fish species based on otolith contour: using short-time Fourier transform and discriminant analysis (STFT-DA)

• DOI: 10.7287/peerj.preprints.1517v1
• Publication Date: 2018-01-12T23:33:00Z
• Authors: Nima Salimi; Kar Hoe Loh; Sarinder Kaur Dhi...; Ving Ching Chong
• Keywords: Identification

Background. Fish species may be identified based on their unique otolith shape or contour. Several pattern recognition methods have been proposed to classify fish through morphological features of the contours. However, there has no fully-automated identification model with accuracy higher than 80%. The purpose current study is develop a model, contours, identify high classification accuracy. Methods. Images right sagittal otoliths 14 from three families namely Sciaenidae, Ariidae, and Engraulidae were used model. Short-time Fourier transform (STFT) was used, for first time in area analysis, extract important Discriminant Analysis (DA), as technique, train test extracted features. Results. Performance demonstrated using separately, well all combined. Overall greater 90% cases. In addition, effects STFT variables performance explored this study. Conclusions. could determine outlines. (STFT-DA) predict an unknown specimen acceptable flexibility more future studies.