DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
Rechercher
(titres de publication, de périodique et noms de colloque inclus)
2017-07-24 - Colloque/Article dans les actes avec comité de lecture - Anglais - 8 page(s)

Florentin Juliette , Verlinden Olivier , "Autonomous Wildlife Soundscape Recording Station Using Raspberry Pi" in Proceedings of the 24th International Congress on Sound and Vibration (ICSV24), London, UK, 2017

  • Codes CREF : Sciences de l'ingénieur (DI2000), Intelligence artificielle (DI1180), Ornithologie (DI3166), Acoustique (DI1264)
  • Unités de recherche UMONS : Mécanique rationnelle, Dynamique et Vibrations (F703)
  • Instituts UMONS : Institut des Biosciences (Biosciences)
Texte intégral :

Abstract(s) :

(Anglais) As acoustic signal processing capabilities progress, it becomes possible to monitor wildlife by recording and analyzing outdoor soundscapes. This is especially true for birds and oth-er vocal animals. The present paper tackles the hardware side and thus describes the com-ponents of an autonomous recording station, intended for winter and spring operation in Belgium (forest environment, negative temperatures, rain, snow, wind). The central ele-ment of the station is a Raspberry Pi computer, to which the following accessories con-nect: an external sound card and omnidirectional microphone, a 3G SIM card dongle (to operate the station remotely), a 64 GB memory stick, a Sleepy Pi module (Raspberry Pi power switch) and two 12 V lead acid batteries. The microphone head is protected from humidity and wind and its body is kept warm inside the sealed electronics box. The Rasp-berry Pi processes the audio stream as it is recorded using GNU Octave. The data is resampled and assessed to save disk space. The Acoustic Complexity Indicator (ACI) sup-plies a) daily summary images that show the wildlife acoustic activity on site and b) a se-lection of interesting sound segments that are saved to disk. The primary challenges are the robustness and the autonomy of the station. For the first item, the Sleepy Pi in combina-tion with proper scripting ensures rebooting of the station in case of failures. For the sec-ond item, battery voltage is tracked and power usage rationalized. The recording station was deployed in 2016 in Wallonia to study the local grey-headed woodpecker (Picus canus) population. It processed over 400 hours of sound and saved 100 hours. 18% of the recorded segments contained woodpecker drums. We collected 2232 woodpecker drums and were able to observe vocal interactions between P. canus and Dryocopus martius (black woodpecker).


Mots-clés :
  • (Anglais) wildlife monitoring
  • (Anglais) bioacoustics
  • (Anglais) recording station
  • (Anglais) ecoacoustics
  • (Anglais) Raspberry Pi