Accurate Multilevel Classification for Wildlife Images
Por favor, use este identificador para citar o enlazar este ítem:
http://hdl.handle.net/10045/114055
Título: | Accurate Multilevel Classification for Wildlife Images |
---|---|
Autor/es: | Gomez-Donoso, Francisco | Escalona, Félix | Pérez Esteve, Ferran | Cazorla, Miguel |
Grupo/s de investigación o GITE: | Robótica y Visión Tridimensional (RoViT) |
Centro, Departamento o Servicio: | Universidad de Alicante. Departamento de Ciencia de la Computación e Inteligencia Artificial | Universidad de Alicante. Instituto Universitario de Investigación Informática |
Palabras clave: | Multilevel classification | Wild animals | Plant species | Wildlife images | Accuracy |
Área/s de conocimiento: | Ciencia de la Computación e Inteligencia Artificial |
Fecha de publicación: | 2-abr-2021 |
Editor: | Hindawi |
Cita bibliográfica: | Computational Intelligence and Neuroscience. Volume 2021, Article ID 6690590, 11 pages. https://doi.org/10.1155/2021/6690590 |
Resumen: | The most common approaches for classification rely on the inference of a specific class. However, every category could be naturally organized within a taxonomic tree, from the most general concept to the specific element, and that is how human knowledge works. This representation avoids the necessity of learning roughly the same features for a range of very similar categories, and it is easier to understand and work with and provides a classification for each abstraction level. In this paper, we carry out an exhaustive study of different methods to perform multilevel classification applied to the task of classifying wild animals and plant species. Different convolutional backbones, data setups, and ensembling techniques are explored to find the model which provides the best performance. As our experimentation remarks, in order to achieve the best performance on the datasets that are arranged in a tree-like structure, the classifier must feature an EfficientNetB5 backbone with an input size of px, followed by a multilevel classifier. In addition, a Multiscale Crop data augmentation process must be carried out. Finally, the accuracy of this setup is a 62% top-1 accuracy and 88% top-5 accuracy. The architecture could benefit for an accuracy boost if it is involved in an ensemble of cascade classifiers, but the computational demand is unbearable for any real application. |
Patrocinador/es: | This work was funded by the Spanish Government PID2019-104818RB-I00 grant, supported with FEDER funds. It was supported by Spanish grants for PhD studies ACIF/2017/243 and FPU16/00887. |
URI: | http://hdl.handle.net/10045/114055 |
ISSN: | 1687-5265 (Print) | 1687-5273 (Online) |
DOI: | 10.1155/2021/6690590 |
Idioma: | eng |
Tipo: | info:eu-repo/semantics/article |
Derechos: | © 2021 Francisco Gomez-Donoso et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. |
Revisión científica: | si |
Versión del editor: | https://doi.org/10.1155/2021/6690590 |
Aparece en las colecciones: | INV - RoViT - Artículos de Revistas |
Archivos en este ítem:
Archivo | Descripción | Tamaño | Formato | |
---|---|---|---|---|
Gomez-Donoso_etal_2021_ComputIntelligNeurosci.pdf | 4,85 MB | Adobe PDF | Abrir Vista previa | |
Este ítem está licenciado bajo Licencia Creative Commons