White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Automated Ladybird Identification using Neural and Expert Systems

Ayob, Mohd Zaki (2012) Automated Ladybird Identification using Neural and Expert Systems. PhD thesis, University of York.

Text (MZA thesis)
Available under License Creative Commons Attribution-Noncommercial-No Derivative Works 2.0 UK: England & Wales.

Download (4Mb)


The concept of automated species identification is relatively recent and advances are being driven by technological advances and the taxonomic impediment. This thesis describes investigations into the automated identification of ladybird species from colour images provided by the public, with an eventual aim of implementing an online identification system. Such images pose particularly difficult problems with regards to image processing as the insects have a highly domed shape and not all relevant features (e.g. spots) are visible or are fore-shortened. A total of 7 species of ladybird have been selected for this work; 6 native species to the UK and 3 colour forms of the Harlequin ladybird (Harmonia axyridis), the latter because of its pest status. Work on image processing utilised 6 geometrical features obtained using greyscale operations, and 6 colour features which were obtained using CIELAB colour space representation. Overall classifier results show that inter-species identification is a success; the system is able to, among all, correctly identify Calvia 14-guttata from Halyzia 16-guttata to 100% accuracy and Exochomus 4-pustulatus from H. axyridis f. spectabilis to 96.3% accuracy using Multilayer Perceptron and J48 decision trees. Intra-species identification of H. axyridis shows that H. axyridis f. spectabilis can be identified correctly up to 72.5% against H. axyridis f. conspicua, and 98.8% correct against H. axyridis f. succinea. System integration tests show that through the addition of user interaction, the identification between Harlequins and non-Harlequins can be improved from 18.8% to 75% accuracy.

Item Type: Thesis (PhD)
Additional Information: This is my final revised thesis, please ignore the 2nd revision (accidental deposit).
Academic Units: The University of York > Electronics (York)
Identification Number/EthosID: uk.bl.ethos.577374
Depositing User: Mr Mohd Zaki Ayob
Date Deposited: 19 Aug 2013 10:53
Last Modified: 24 Jul 2018 15:20
URI: http://etheses.whiterose.ac.uk/id/eprint/4290

You do not need to contact us to get a copy of this thesis. Please use the 'Download' link(s) above to get a copy.
You can contact us about this thesis. If you need to make a general enquiry, please see the Contact us page.

Actions (repository staff only: login required)