%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2017/08.16.12.58
%2 sid.inpe.br/sibgrapi/2017/08.16.12.58.57
%@doi 10.1109/SIBGRAPI.2017.32
%T Drosophila melanogaster Gender Classification Based on Fractal Dimension
%D 2017
%A Medeiros Neto, Francisco Gerardo,
%A Braga, Ítalo Rodrigues,
%A Harber, Matthew Henry,
%A Júnior, Iális Cavalcante de Paula,
%@affiliation Federal University of Ceará
%@affiliation Federal University of Ceará
%@affiliation GeoPoll
%@affiliation Federal University of Ceará
%E Torchelsen, Rafael Piccin,
%E Nascimento, Erickson Rangel do,
%E Panozzo, Daniele,
%E Liu, Zicheng,
%E Farias, Mylčne,
%E Viera, Thales,
%E Sacht, Leonardo,
%E Ferreira, Nivan,
%E Comba, Joăo Luiz Dihl,
%E Hirata, Nina,
%E Schiavon Porto, Marcelo,
%E Vital, Creto,
%E Pagot, Christian Azambuja,
%E Petronetto, Fabiano,
%E Clua, Esteban,
%E Cardeal, Flávio,
%B Conference on Graphics, Patterns and Images, 30 (SIBGRAPI)
%C Niterói, RJ, Brazil
%8 17-20 Oct. 2017
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K stationary wavelet transform, Canny filter, fractal dimension, classification.
%X Biometrics, previously used only in human identification, can help experts in the analysis of biological images. Flies of the genus Drosophila have become model organisms by almost global presence and short life cycle. Facial recognition techniques and geometric morphometry can be used in image processing for classification. The latter requires human interaction. This work details a methodology based on stationary wavelet transform, Canny filter and fractal dimension aimed to infer the gender of Drosophila melanogaster based on images of their wings. The combination of variation in the training and test samples and classification methods showed the proposed algorithms accuracy rate, 90%, outperformed other methods. The proposed methodology proved efficient by using a reduced number of attributes and did not require human interaction for feature extraction (landmarks).
%@language en
%3 sibgrapi-2017-cr.pdf