%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2018/10.22.14.04
%2 sid.inpe.br/sibgrapi/2018/10.22.14.04.11
%T Evaluation of convolutional neural networks for raw food texture classification under variations of lighting conditions
%D 2018
%A Ferraz, Carolina Toledo,
%A Borges, Tamiris T. N.,
%A Cavichiolli, Adriane,
%A Gonzaga, Adilson,
%A Saito, José H.,
%@affiliation UNIFACCAMP
%@affiliation Federal Institute of São Paulo
%@affiliation University of São Paulo
%@affiliation University of São Paulo
%@affiliation UNIFACCAMP
%E Ross, Arun,
%E Gastal, Eduardo S. L.,
%E Jorge, Joaquim A.,
%E Queiroz, Ricardo L. de,
%E Minetto, Rodrigo,
%E Sarkar, Sudeep,
%E Papa, João Paulo,
%E Oliveira, Manuel M.,
%E Arbeláez, Pablo,
%E Mery, Domingo,
%E Oliveira, Maria Cristina Ferreira de,
%E Spina, Thiago Vallin,
%E Mendes, Caroline Mazetto,
%E Costa, Henrique Sérgio Gutierrez,
%E Mejail, Marta Estela,
%E Geus, Klaus de,
%E Scheer, Sergio,
%B Conference on Graphics, Patterns and Images, 31 (SIBGRAPI)
%C Foz do Iguaçu, PR, Brazil
%8 29 Oct.-1 Nov. 2018
%I Sociedade Brasileira de Computação
%J Porto Alegre
%S Proceedings
%K texture classification, CNN, light intensity.
%X This work is a preliminary evaluation of convolutional neural networks (CNN) applied to food texture classification, particularly when the texture is subject to changes in the lighting conditions. Four previously published CNN architectures (Alexnet, Resnet 18, Resnet 34 and Resnet 50) are investigated and compared to local descriptors designed specifically for this task. Although preliminary results indicate that the investigated CNN are outperformed by the descriptors, further analysis are required to investigate the impact of the experimental design adopted in this work-in-progress; especially in regard to the number of training samples and CNN configuration.
%@language en
%3 sibgrapi_2018_versaofinal.pdf