Keywords

Otoliths, Feature space, Classifiers, Automatic Taxon Identification (ATI), Elliptic Fourier Descriptors (EFD), Pattern Matching, Pre-processing, Data mining

Location

Session G2: Data Mining for Environmental Sciences

Start Date

17-6-2014 9:00 AM

End Date

17-6-2014 10:20 AM

Abstract

This paper analyzes the characteristics of a rotation-invariant Feature space to be used in a classifier of fish otoliths, it is compared to two other Feature spaces, one with raw data and another with transformed data (using the Elliptic Fourier Descriptors EFD). Otoliths are found in the inner ear of fish. Their shape can be analyzed to determine sex, age, populations and species, and thus they can provide necessary and relevant information for ecological studies. The Automatic Taxon Identifier (ATI) is used to classify fish otoliths directly from a query image and is implemented on-line in a Public Database. This new Features space proposed in this study will allow the use of query images that are not normalized or are normalized in a different pose than the images in the Database. The rotation-invariant Feature space is the pre-processing part of a classifier based in kNearestNeighbours, and it is derived from the Elliptical Fourier Descriptors (EFD). The new Feature space compacts the information, and could be used when a feature reduction strategy is recommended. The authors analyze the classification results in a Test Database applying a controlled rotation to the Query images, and discuss their possible utility of the technique in the future ATI system.

Share

COinS
 
Jun 17th, 9:00 AM Jun 17th, 10:20 AM

Pre-processing techniques applied to Automatic Taxon Identification on fish otoliths

Session G2: Data Mining for Environmental Sciences

This paper analyzes the characteristics of a rotation-invariant Feature space to be used in a classifier of fish otoliths, it is compared to two other Feature spaces, one with raw data and another with transformed data (using the Elliptic Fourier Descriptors EFD). Otoliths are found in the inner ear of fish. Their shape can be analyzed to determine sex, age, populations and species, and thus they can provide necessary and relevant information for ecological studies. The Automatic Taxon Identifier (ATI) is used to classify fish otoliths directly from a query image and is implemented on-line in a Public Database. This new Features space proposed in this study will allow the use of query images that are not normalized or are normalized in a different pose than the images in the Database. The rotation-invariant Feature space is the pre-processing part of a classifier based in kNearestNeighbours, and it is derived from the Elliptical Fourier Descriptors (EFD). The new Feature space compacts the information, and could be used when a feature reduction strategy is recommended. The authors analyze the classification results in a Test Database applying a controlled rotation to the Query images, and discuss their possible utility of the technique in the future ATI system.