Assemblage of Focal Species Recognizers—AFSR: A technique for decreasing false indications of presence from acoustic automatic identification in a multiple species context
Autoři:
Ivan Braga Campos aff001; Todd J. Landers aff001; Kate D. Lee aff001; William George Lee aff001; Megan R. Friesen aff001; Anne C. Gaskett aff001; Louis Ranjard aff005
Působiště autorů:
Centre for Biodiversity and Biosecurity, School of Biological Sciences, University of Auckland, Auckland, New Zealand
aff001; Chico Mendes Institute for Biodiversity Conservation, Serra do Cipó National Park, Serra do Cipó/MG, Brasil
aff002; Research and Evaluation Unit, Auckland Council, Auckland, New Zealand
aff003; Landcare Research, Dunedin, New Zealand
aff004; Research School of Biology, ANU College of Medicine, Biology and Environment, The Australian National University, Canberra, ACT, Australia
aff005
Vyšlo v časopise:
PLoS ONE 14(12)
Kategorie:
Research Article
doi:
https://doi.org/10.1371/journal.pone.0212727
Souhrn
Passive acoustic monitoring (PAM) coupled with automated species identification is a promising tool for species monitoring and conservation worldwide. However, high false indications of presence are still an important limitation and a crucial factor for acceptance of these techniques in wildlife surveys. Here we present the Assemblage of Focal Species Recognizers—AFSR, a novel approach for decreasing false positives and increasing models’ precision in multispecies contexts. AFSR focusses on decreasing false positives by excluding unreliable sound file segments that are prone to misidentification. We used MatlabHTK, a hidden Markov models interface for bioacoustics analyses, for illustrating AFSR technique by comparing two approaches, 1) a multispecies recognizer where all species are identified simultaneously, and 2) an assemblage of focal species recognizers (AFSR), where several recognizers that each prioritise a single focal species are then summarised into a single output, according to a set of rules designed to exclude unreliable segments. Both approaches (the multispecies recognizer and AFSR) used the same sound files training dataset, but different processing workflow. We applied these recognisers to PAM recordings from a remote island colony with five seabird species and compared their outputs with manual species identifications. False positives and precision improved for all the five species when using AFSR, achieving remarkable 0% false positives and 100% precision for three of five seabird species, and < 6% false positives, and >90% precision for the other two species. AFSR’ output was also used to generate daily calling activity patterns for each species. Instead of attempting to withdraw useful information from every fragment in a sound recording, AFSR prioritises more trustworthy information from sections with better quality data. AFSR can be applied to automated species identification from multispecies PAM recordings worldwide.
Klíčová slova:
Acoustics – Bioacoustics – Birds – Hidden Markov models – Islands – Seabirds – Petrels – Bird song
Zdroje
1. Zilli D, Parson O, Merrett GV, Rogers A. A hidden Markov model-based acoustic cicada detector for crowdsourced smartphone biodiversity monitoring. Journal of Artificial Intelligence Research 2014;51:805–827.
2. Aide TM, Corrada-Bravo C, Campos-Cerqueira M, Milan C, Vega G, Alvarez R. Real-time bioacoustics monitoring and automated species identification. PeerJ 2013;1(e103):1–19.
3. Xie J, Towsey M, Zhang L, Yasumiba K, Schwarzkopf L, Zhang J, et al. Multiple-Instance Multiple-Label Learning for the Classification of Frog Calls with Acoustic Event Detection. In: Mansouri A, Nouboud F, Chalifour A, Mammass D, Meunier J, Elmoataz A, editors. Image and Signal Processing: 7th International Conference, ICISP 2016, Trois-Rivières, QC, Canada, May 30—June 1, 2016, Proceedings Cham: Springer International Publishing; 2016. p. 222–230.
4. Deichmann JL, Hernández-Serna A, Campos-Cerqueira M, Aide TM. Soundscape analysis and acoustic monitoring document impacts of natural gas exploration on biodiversity in a tropical forest. Ecol Ind 2017;74:39–48.
5. Andreassen T, Surlykke A, Hallam J. Semi-automatic long-term acoustic surveying: A case study with bats. Ecological Informatics 2014;21:13–24.
6. Newson SE, Evans HE, Gillings S. A novel citizen science approach for large-scale standardised monitoring of bat activity and distribution, evaluated in eastern England. Biol Conserv 2015;191:38–49.
7. Rocha LH, Ferreira LS, Paula BC, Rodrigues FH, Sousa-Lima RS. An evaluation of manual and automated methods for detecting sounds of maned wolves (Chrysocyon brachyurus Illiger 1815). Bioacoustics 2015;24(2):185–198.
8. Palacios V, López-Bao JV, Llaneza L, Fernández C, Font E. Decoding Group Vocalizations: The Acoustic Energy Distribution of Chorus Howls Is Useful to Determine Wolf Reproduction. PloS one 2016;11(5):e0153858. doi: 10.1371/journal.pone.0153858 27144887
9. Sanders CE, Mennill DJ. Acoustic monitoring of nocturnally migrating birds accurately assesses the timing and magnitude of migration through the Great Lakes. The Condor 2014;116(3):371–383.
10. Stowell D, Benetos E, Gill LF. On-bird Sound Recordings: Automatic Acoustic Recognition of Activities and Contexts. IEEE/ACM Transactions on Audio, Speech, And Language Processing 2017;25(6):1193–1206.
11. Ranjard L, Withers SJ, Brunton D, Ross HA, Parsons S. Integration over song classification replicates: evidence for song types and micro-geographic variation in the hihi. Journal of the Acoustical Society of America 2015;137(5):2542–2551. doi: 10.1121/1.4919329 25994687
12. Ranjard L, Withers SJ, Brunton DH, Parsons S, Ross HA. Geographic patterns of song variation reveal timing of song acquisition in a wild avian population. Behavioral Ecology 2017;28(4):1085–1092.
13. Putland RL, Ranjard L, Constantine R, Radford CA. A hidden Markov model approach to indicate Bryde’s whale acoustics. Ecological Indicators 2018;84(Supplement C):479–487.
14. Wrege PH, Rowland ED, Keen S, Shiu Y. Acoustic monitoring for conservation in tropical forests: examples from forest elephants. Methods in Ecology and Evolution 2017;8(10):1292–1301.
15. Cragg JL, Burger AE, Piatt JF. Testing the effectiveness of automated acoustic sensors for monitoring vocal activity of marbled murrelets Brachyramphus marmoratus. Mar Ornithol 2015;43:151–160.
16. Zwart MC, Baker A, McGowan PJK, Whittingham MJ. The Use of Automated Bioacoustic Recorders to Replace Human Wildlife Surveys: An Example Using Nightjars. PLOS ONE 2014;9(7):e102770. doi: 10.1371/journal.pone.0102770 25029035
17. Borker AL, McKown MW, Ackerman JT, EAGLES‐SMITH CA, Tershy BR, Croll DA. Vocal activity as a low cost and scalable index of seabird colony size. Conserv Biol 2014;28(4):1100–1108. doi: 10.1111/cobi.12264 24628442
18. Jennings N, Parsons S, Pocock M. Human vs. machine: identification of bat species from their echolocation calls by humans and by artificial neural networks. Can J Zool 2008;86(5):371–377.
19. Buxton RT, Jones IL. Measuring nocturnal seabird activity and status using acoustic recording devices: applications for island restoration. J Field Ornithol 2012;83(1):47–60.
20. Mulder CPH, Anderson WB, Towns DR, Bellingham PJ editors. Seabird islands: ecology, invasion, and restoration. New York: Oxford University Press; 2011.
21. Potamitis I, Ntalampiras S, Jahn O, Riede K. Automatic bird sound detection in long real-field recordings: Applications and tools. Appl Acoust 2014;80:1–9.
22. Ranjard L, Reed BS, Landers TJ, Rayner MJ, Friesen MR, Sagar RL, et al. MatlabHTK: a simple interface for bioacoustic analyses using hidden Markov models. Methods in Ecology and Evolution 2017;8:615–621.
23. Ismar SMH, Baird KA, Gaskin CP, Taylor GA, Tennyson AJD, Rayner MJ, et al. A case of natural recovery after the removal of invasive predators—community assemblage changes in the avifauna of Burgess Island. Notornis 2014;61:188–195.
24. Young SJ, Evermann G, Gales MJF, Hain T, Kershaw D, Moore G, et al. The HTK Book. Cambridge University Engineering Department, Cambridge 2006.
25. Eaton JW, Bateman D, Hauberg S, and Wehbring R. GNU Octave version 3.8.1 manual: a high-level interactive language for numerical computations. 2014.
26. Gage SH, Napoletano BM, Cooper MC. Assessment of ecosystem biodiversity by acoustic diversity indices. J Acoust Soc Am 2001;109(5):2430.
27. Warham J. The petrels: their ecology and breeding systems. London: Academic Press; 1990.
28. Landers TJ, Bannock CA, Hauber ME. Dynamics of behavioural rhythms in a colonial, nocturnal, burrowing seabird: a comparison across different temporal scales. Notornis 2011;58:81–89.
29. Ross EL, Brunton D. Seasonal trends and nightly variation in colony attendance of grey-faced petrels (Pterodroma macroptera gouldi). Notornis 2002;49(3):153–157.
30. Howald G, Donlan C, Galván JP, Russell JC, Parkes J, Samaniego A, et al. Invasive rodent eradication on islands. Conserv Biol 2007;21(5):1258–1268. doi: 10.1111/j.1523-1739.2007.00755.x 17883491
31. Bellingham PJ, Towns DR, Cameron EK, Davis JJ, Wardle DA, Wilmshurst JM, et al. New Zealand island restoration: seabirds, predators, and the importance of history. N Z J Ecol 2010;34(1):115.
Článek vyšel v časopise
PLOS One
2019 Číslo 12
- S diagnostikou Parkinsonovy nemoci může nově pomoci AI nástroj pro hodnocení mrkacího reflexu
- Je libo čepici místo mozkového implantátu?
- Pomůže v budoucnu s triáží na pohotovostech umělá inteligence?
- AI může chirurgům poskytnout cenná data i zpětnou vazbu v reálném čase
- Nová metoda odlišení nádorové tkáně může zpřesnit resekci glioblastomů
Nejčtenější v tomto čísle
- Methylsulfonylmethane increases osteogenesis and regulates the mineralization of the matrix by transglutaminase 2 in SHED cells
- Oregano powder reduces Streptococcus and increases SCFA concentration in a mixed bacterial culture assay
- The characteristic of patulous eustachian tube patients diagnosed by the JOS diagnostic criteria
- Parametric CAD modeling for open source scientific hardware: Comparing OpenSCAD and FreeCAD Python scripts
Zvyšte si kvalifikaci online z pohodlí domova
Všechny kurzy