2018
DOI: 10.1007/s10994-018-5739-8
|View full text |Cite
|
Sign up to set email alerts
|

A comparison of hierarchical multi-output recognition approaches for anuran classification

Abstract: In bioacoustic recognition approaches, a "flat" classifier is usually trained to recognize several species of anurans, where the number of classes is equal to the number of species. Consequently, the complexity of the classification function increases proportionally with the number of species. To avoid this issue, we propose a "hierarchical" approach that decomposes the problem into three taxonomic levels: the family, the genus, and the species. To accomplish this, we transform the original single-labelled pro… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
12
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
6
2
1

Relationship

2
7

Authors

Journals

citations
Cited by 17 publications
(12 citation statements)
references
References 34 publications
0
12
0
Order By: Relevance
“…Classical tools such as confusion matrices are also used as interpretation tools on the way to scientific outcomes. In a bio-acoustic application for the recognition of anurans using acoustic sensors, Colonna et al [2018] use a hierarchical approach to jointly classify on three taxonomic levels, namely the family, the genus, and the species. Investigating the confusion matrix per level enabled for example the identification of bio-acoustic similarities between different species.…”
Section: Interpretation Tools For Scientific Outcomesmentioning
confidence: 99%
“…Classical tools such as confusion matrices are also used as interpretation tools on the way to scientific outcomes. In a bio-acoustic application for the recognition of anurans using acoustic sensors, Colonna et al [2018] use a hierarchical approach to jointly classify on three taxonomic levels, namely the family, the genus, and the species. Investigating the confusion matrix per level enabled for example the identification of bio-acoustic similarities between different species.…”
Section: Interpretation Tools For Scientific Outcomesmentioning
confidence: 99%
“…To reduce the risk of false positives and false negatives (especially when dealing with unknown species), hierarchical classification approaches could be developed. These methods are a known technique to improve model generalization and have been shown to be relevant in handling biological data (Colonna, Gama, & Nakamura, ; Redmon & Farhadi, ).…”
Section: Discussionmentioning
confidence: 99%
“…A way to handle these issues can be an implementation of a hierarchical classification. Such algorithms have already been successfully used in general object detection methods [13] as well as for handling ecological data [14].…”
Section: Discussionmentioning
confidence: 99%