Bayesian Classification of Flight Calls with a Novel Dynamic Time Warping Kernel

Damoulas, Theodoros; Henry, Sam; Farnsworth, Andrew; Lanzone, Michael; Gomes, Carla P.

doi:10.1109/icmla.2010.69

Cited by 15 publications

(15 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…; Damoulas et al . ) and most bat acoustic classification tasks still represent classifications with a few parameters and further classify them using manual or nonparametric techniques. Such whole signal analyses in bat acoustics are growing (Obrist, Boesch & Flückiger ; Skowronski & Harris ; Stathopoulos et al .…”

Section: Discussionmentioning

confidence: 99%

“…However, applications of these approaches have mainly focused on bird and marine mammal acoustics (e.g. Ren et al 2009;Damoulas et al 2010) and most bat acoustic classification tasks still represent classifications with a few parameters and further classify them using manual or nonparametric techniques. Such whole signal analyses in bat acoustics are growing (Obrist, Boesch & Fl€ uckiger 2004;Skowronski & Harris 2006;Stathopoulos et al 2014) but should be further explored.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design

et al. 2016

View full text Add to dashboard Cite

Summary Monitoring global biodiversity is critical for understanding responses to anthropogenic change, but biodiversity monitoring is often biased away from tropical, megadiverse areas that are experiencing more rapid environmental change. Acoustic surveys are increasingly used to monitor biodiversity change, especially for bats as they are important indicator species and most use sound to detect, localise and classify objects. However, using bat acoustic surveys for monitoring poses several challenges, particularly in megadiverse regions. Many species lack reference recordings, some species have high call similarity or differ in call detectability, and quantitative classification tools, such as machine learning algorithms, have rarely been applied to data from these areas. Here, we collate a reference call library for bat species that occur in a megadiverse country, Mexico. We use 4685 search‐phase calls from 1378 individual sequences of 59 bat species to create automatic species identification tools generated by machine learning algorithms (Random Forest). We evaluate the improvement in species‐level classification rates gained by using hierarchical classifications, reflecting either taxonomic or ecological constraints (guilds) on call design, and examine how classification rate accuracy changes at different hierarchical levels (family, genus and guild). Species‐level classification of calls had a mean accuracy of 66%, and the use of hierarchies improved mean species‐level classification accuracy by up to 6% (species within families 72%, species within genera 71·2% and species within guilds 69·1%). Classification accuracy to family, genus and guild‐level was 91·7%, 77·8% and 82·5%, respectively. The bioacoustic identification tools we have developed are accurate for rapid biodiversity assessments in a megadiverse region and can also be used effectively to classify species at broader taxonomic or ecological levels. This flexibility increases their usefulness when there are incomplete species reference recordings and also offers the opportunity to characterise and track changes in bat community structure. Our results show that bat bioacoustic surveys in megadiverse countries have more potential than previously thought to monitor biodiversity changes and can be used to direct further developments of bioacoustic monitoring programs in Mexico.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design

et al. 2016

View full text Add to dashboard Cite

show abstract

“…However, owing to the differences in call duration the spectrograms will need to be normalized to have the same length by using some form of interpolation. In this work we borrow ideas from speech recognition (Sakoe and Chiba, 1978) and previous work on bird call classification (Damoulas et al, 2010) and employ the dynamic time warping (DTW) kernel to compare two calls' spectrograms directly.…”

Section: Signal Processing and Data Representationmentioning

confidence: 99%

Bat Echolocation Call Identification for Biodiversity Monitoring: A Probabilistic Approach

Stathopoulos

Zamora‐Gutiérrez

Jones

et al. 2017

Journal of the Royal Statistical Society Series C: Applied Statistics

View full text Add to dashboard Cite

Summary. Bat echolocation call identification methods are important in developing efficient cost-effective methods for large-scale bioacoustic surveys for global biodiversity monitoring and conservation planning. Such methods need to provide interpretable probabilistic predictions of species since they will be applied across many different taxa in a diverse set of applications and environments. We develop such a method using a multinomial probit likelihood with independent Gaussian process priors and study its feasibility on a data set from an on-going study of 21 species, five families and 1800 bat echolocation calls collected from Mexico, a hotspot of bat biodiversity. We propose an efficient approximate inference scheme based on the expectation propagation algorithm and observe that the overall methodology significantly improves on currently adopted approaches to bat call classification by providing an approach which can be easily generalized across different species and call types and is fully probabilistic. Implementation of this method has the potential to provide robust species identification tools for biodiversity acoustic bat monitoring programmes across a range of taxa and spatial scales.

show abstract

“…Because flight calls are most common during migration, acoustic records can help describe inflight migrant species composition, which is unavailable from other standard methods like radar, thermal imaging, moon-watching, etc. Of the tools available to monitor active migration, our understanding of these unique calls is least understood (Farnsworth, 2005;Keen, Ross, Griffiths, Lanzone, & Farnsworth, 2014) With the current library of North American flight calls exceeding 200 species, and because acoustic data are both easy and inexpensive to collect, the potential for networks of acoustic monitoring stations to provide comprehensive coverage of nocturnal migrants is great (Damoulas, Henry, Farnsworth, Lanzone, & Gomes, 2010). Acoustic monitoring can capture instantaneous movements of species, lending acute resolution to questions of migratory phenology and distribution.…”

Section: Introductionmentioning

confidence: 99%

Individual flight-calling behaviour in wood warblers

et al. 2016

Self Cite

View full text Add to dashboard Cite

Bayesian Classification of Flight Calls with a Novel Dynamic Time Warping Kernel

Cited by 15 publications

References 21 publications

Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design

Acoustic identification of Mexican bats based on taxonomic and ecological constraints on call design

Bat Echolocation Call Identification for Biodiversity Monitoring: A Probabilistic Approach

Individual flight-calling behaviour in wood warblers

Contact Info

Product

Resources

About