Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.500781
Title: Intelligent visual otolith classification for bony fish species recognition
Author: Lefkaditis, Dionysios
Awarding Body: University of Brighton
Current Institution: University of Brighton
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Abstract:
The study of otoliths is a well-established source of information for understanding the life offish and fish populations. Conducting fish species identification from otolith samples found in the stomach contents of marine fish-eating animals finds interesting applications such as dietary studies, stock monitoring, assessment and management. Fish species identification can provide useful data for climatology, archaeology and palaeontology research, as otoliths can be sourced from geological sediments or archaeological excavations. Analysing an otolith is a highly complex and time-consuming procedure Therefore, an automated otolith classification system can prove to be a vital tool for a wide variety of scientific research. The aim of the programme of work seeks the development of a novel automated fish species identification system. The main focus of this investigation is on the commercially interesting fish of the Northern Aegean Sea. The methodology described in this thesis exploits the inherent shape variability offish otoliths according to their corresponding species. This is based on the processing and analysis of images acquired using a stereoscopic microscope fitted with a digital camera. A compact feature vector is then constructed out of a list of candidate descriptors derived from the morphology as well as the image statistics of the otoliths. The identification is carried out by an intelligent classifier based on an artificial neural network. Several configurations of multi-layer perceptron, radial basis function and hybrid neural networks are considered in pursuit of a practical and expandable classification system.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.500781  DOI: Not available
Share: