Soundscape segregation based on visual analysis and discriminating features

Dias, Fábio Felix; Pedrini, Hélio; Minghim, Rosane

doi:10.1016/j.ecoinf.2020.101184

Cited by 15 publications

(17 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The resulting graphics, known as spectrograms, reveal patterns in both frequency content of a soundscape as well as temporal patterns. While the resolution of these graphics is user defined, these visualization techniques are a means of data averaging and result in long-term spectral averages that can be used to visualize unique characteristics of a site or presence of acoustic events (Dias et al, 2021). These visual representations of sound are valuable for soundscape exploration, qualitative descriptions, and manual extraction of sound events; additional analytical steps are typically necessary to derive quantitative and comparable metrics for sound levels.…”

Section: Current Approaches To Characterization Of Marine Soundscapesmentioning

confidence: 99%

“…In some cases, classification was derived from band-level indices (e.g., Širović et al, 2015;Oestreich et al, 2020). When the training data are insufficient, additional exemplars can be added by transforming the true signals (and labeling them as false) or adding different types and amounts (Barchiesi et al, 2015;Benetos et al, 2015;Shane et al, 2015;Zabalza et al, 2015;Freeman and Freeman, 2016;Lin et al, 2017a,b;Lin et al, , 2019Lin et al, , 2021Tsao, 2018, 2020;Colonna et al, 2018;Guan et al, 2018;Reis et al, 2018;Seger et al, 2018;Virtanen et al, 2018;Abeßer, 2020;Gatto et al, 2020;Mooney et al, 2020;Shajahan et al, 2020;Xie et al, 2020;Dias et al, 2021;Dimoff et al, 2021;Gabrielli et al, 2021;Ick and McFee, 2021;Ozanich et al, 2021).…”

Section: Analytical Approaches To Soundscape Analysesmentioning

confidence: 99%

“…The literature referenced simply serve as examples, and not an exhaustive or systematic search of the literature. References cited in the table:(Barchiesi et al, 2015;Benetos et al, 2015;Shane et al, 2015;Zabalza et al, 2015;Freeman and Freeman, 2016; Lin et al, 2017a,b;Lin et al, , 2019Lin et al, , 2021 Tsao, 2018, 2020;Colonna et al, 2018;Guan et al, 2018;Reis et al, 2018;Seger et al, 2018;Virtanen et al, 2018;Abeßer, 2020;Gatto et al, 2020;Mooney et al, 2020;Shajahan et al, 2020;Xie et al, 2020;Dias et al, 2021;Dimoff et al, 2021;Gabrielli et al, 2021;Ick and McFee, 2021;Ozanich et al, 2021).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Advancing the Interpretation of Shallow Water Marine Soundscapes

et al. 2021

View full text Add to dashboard Cite

Soundscapes offer rich descriptions of composite acoustic environments. Characterizing marine soundscapes simply through sound levels results in incomplete descriptions, limits the understanding of unique features, and impedes meaningful comparisons. Sources that contribute to sound level metrics shift in time and space with changes in biological patterns, physical forces, and human activity. The presence of a constant or chronic source is often interwoven with episodic sounds. Further, the presence and intensity of sources can influence other sources, calling for a more integrated approach to characterizing soundscapes. Here, we illustrate this complexity using data from a national-scale effort, the Sanctuary Soundscape Monitoring Project (SanctSound), an initiative designed to support collection of biological, environmental, and human use data to compliment the interpretation of sound level measurements. Using nine examples from this diverse dataset we demonstrate the benefit of integrating source identification and site features to interpret sound levels across a diversity of shallow water marine soundscapes (<150 m). Sound levels from sites in high human use areas reflect the anthropogenic influences on the soundscape, especially when measuring broad frequency bands, whereas sites with relatively low human activity and high sound levels reveal biological features of the soundscape. At sites with large tidal changes, sound levels correlated with the magnitude of tidal flow, and sound levels during high tidal flow periods were similar to sound levels at sites near commercial shipping lanes. At sites in relatively close proximity (<30 km), sound levels diverge because of the presence of different proximate sound contributors and propagation features of the site. A review of emerging methodologies for integrated soundscape analysis, including acoustic scenes, provides a framework for interpreting soundscapes across a variety of conditions. With a global growth in monitoring efforts collecting standardized measurements over widely distributed arrays, more integrated methods are needed to advance the utility of soundscapes in marine resource management.

show abstract

Section: Current Approaches To Characterization Of Marine Soundscapesmentioning

confidence: 99%

Section: Analytical Approaches To Soundscape Analysesmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Advancing the Interpretation of Shallow Water Marine Soundscapes

et al. 2021

View full text Add to dashboard Cite

show abstract

“…With that transition, ecologists and data scientists are now applying a multitude of data mining tools to the analysis of massive acoustic data. These include those that classify sounds (e.g., Zhao et al, 2017), sort sounds through clustering algorithms (e.g., Bellisario et al, 2019a;Bellisario et al, 2019b), reduce the massive number of acoustic features that are calculated per recording in order to reduce the multidimensionality for more efficient and less complex analysis (Dias et al, 2021;Hilasaca et al, 2021), use of acoustic recordings that are integrated with human perception data (e.g., Aletta et al, 2016) and the development and application of advanced visualization tools such as false color spectrograms (Figure 2). Software development that supports the collection, modification, analysis, fusion, and visualization of acoustic data is needed to advance acoustic remote sensing research.…”

Section: Grand Challengesmentioning

confidence: 99%

Grand Challenges in Acoustic Remote Sensing: Discoveries to Support a Better Understanding of Our Changing Planet

Pijanowski

Brown

2022

Front. Remote Sens.

View full text Add to dashboard Cite

model of the seafloor derived from multibeam bathymetric measurements. Inset image shows how MBES data sets are collected. Imagery courtesy of the Dalhousie University, Seascape Ecology and Mapping Lab (www.seafloormapping.ca); (B) Synthetic Aperture Sonra data set showing a ship wreck and seafloor sediment bedforms. Data is 3 cm horizontal resolution and collected from a towed SAS system shown inset (Kraken Robotics Katfish). Imagery courtesy of Kraken Robotics (https://krakenrobotics.com/); (C) Sub-bottom profiler data showing sediment stratigraphy beneath the seafloor. The profile was collected using a towed sensor platform (Geoforce DTS). Imagery courtesy of Geoforce Group Ltd. (https://www.geoforcegroup.com/).

show abstract

“…In ecological analysis via audio data, for example, as presented by Dias et al [1], Dröge et al [2], Gan et al [3], Hilasaca et al [4], Scarpelli et al [5], acoustic features have an important role in the task of summarizing, representing, visualizing, and analyzing soundscapes. Therefore, removing noise from an audio signal is an important pre-processing step to extract and analyze proper acoustic features related to soundscape dynamics.…”

Section: Introductionmentioning

confidence: 99%