Ant genera identification using an ensemble of convolutional neural networks

Author(s)

Marques, Alan Caio R. ^[1] ; Raimundo, Marcos M. ^[1] ; Cavalheiro, Ellen Marianne B. ^[1] ; Salles, Luis F. P. ^[2] ; Lyra, Christiano ^[1] ; Von Zuben, Fernando J. ^[1]

Affiliation

^[1] Univ Estadual Campinas, UNICAMP, Sch Elect & Comp Engn, Av Albert Einstein 400, BR-13083852 Campinas, SP - Brazil

^[2] Univ Estadual Campinas, UNICAMP, Inst Biol, Grad Program Ecol, R Monteiro Lobato 255, BR-13083862 Campinas, SP - Brazil

• Document type: Journal article
• Source: PLoS One; v. 13, n. 1 JAN 31 2018.
• Web of Science Citations: 3

Abstract

Works requiring taxonomic knowledge face several challenges, such as arduous identification of many taxa and an insufficient number of taxonomists to identify a great deal of collected organisms. Machine learning tools, particularly convolutional neural networks (CNNs), are then welcome to automatically generate high-performance classifiers from available data. Supported by the image datasets available at the largest online database on ant biology, the AntWeb (www.antweb.org), we propose here an ensemble of CNNs to identify ant genera directly from the head, profile and dorsal perspectives of ant images. Transfer learning is also considered to improve the individual performance of the CNN classifiers. The performance achieved by the classifiers is diverse enough to promote a reduction in the overall classification error when they are combined in an ensemble, achieving an accuracy rate of over 80% on top-1 classification and an accuracy of over 90% on top-3 classification. (AU)

• FAPESP's process: 14/13533-0 - Multi-objective optimization in multi-task learning
• Grantee: Marcos Medeiros Raimundo
• Support Opportunities: Scholarships in Brazil - Doctorate