Limits...
A vocal-based analytical method for goose behaviour recognition.

Steen KA, Therkildsen OR, Karstoft H, Green O - Sensors (Basel) (2012)

Bottom Line: A wide range of devices to detect and deter animals causing conflict are used for this purpose, although their effectiveness is often highly variable, due to habituation to disruptive or disturbing stimuli.The Support Vector Machine has proven to be a robust classifier for this kind of classification, as generality and non-linear capabilities are important.We conclude that vocalisations can be used to automatically detect behaviour of conflict wildlife species, and as such, may be used as an integrated part of a wildlife management system.

View Article: PubMed Central - PubMed

Affiliation: Department of Engineering, Aarhus University, Aarhus N, Denmark. kima.steen@agrsci.dk

ABSTRACT
Since human-wildlife conflicts are increasing, the development of cost-effective methods for reducing damage or conflict levels is important in wildlife management. A wide range of devices to detect and deter animals causing conflict are used for this purpose, although their effectiveness is often highly variable, due to habituation to disruptive or disturbing stimuli. Automated recognition of behaviours could form a critical component of a system capable of altering the disruptive stimuli to avoid this. In this paper we present a novel method to automatically recognise goose behaviour based on vocalisations from flocks of free-living barnacle geese (Branta leucopsis). The geese were observed and recorded in a natural environment, using a shielded shotgun microphone. The classification used Support Vector Machines (SVMs), which had been trained with labeled data. Greenwood Function Cepstral Coefficients (GFCC) were used as features for the pattern recognition algorithm, as they can be adjusted to the hearing capabilities of different species. Three behaviours are classified based in this approach, and the method achieves a good recognition of foraging behaviour (86-97% sensitivity, 89-98% precision) and a reasonable recognition of flushing (79-86%, 66-80%) and landing behaviour(73-91%, 79-92%). The Support Vector Machine has proven to be a robust classifier for this kind of classification, as generality and non-linear capabilities are important. We conclude that vocalisations can be used to automatically detect behaviour of conflict wildlife species, and as such, may be used as an integrated part of a wildlife management system.

No MeSH data available.


Related in: MedlinePlus

Sketch of the equipment used for data collection. The camera captures a video stream for later inspection of behaviour. Both audio and video data are stored on an external hard drive. The dashed lines indicate microphone range and camera field of view.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3376600&req=5

f2-sensors-12-03773: Sketch of the equipment used for data collection. The camera captures a video stream for later inspection of behaviour. Both audio and video data are stored on an external hard drive. The dashed lines indicate microphone range and camera field of view.

Mentions: A combination of a shielded shotgun microphone (Sennheiser MKE 400) and a machine vision camera (uEye UI-1245LE-C) with a field of view (FOV) of 45° connected to a laptop were used for recordings. A multiple-shielded audio extension cable was used to minimise loss in fidelity. The camera and laptop were placed in a box at the edge of the field, whereas the microphone was placed 10 m in front of the camera, closer to the geese. The system was powered by two 12 V 92 Ah deep cycling car batteries and data were stored on 3 TB external hard drive. An overview is seen in Figure 2 (a detailed description can be found in Steen et al. [30]).


A vocal-based analytical method for goose behaviour recognition.

Steen KA, Therkildsen OR, Karstoft H, Green O - Sensors (Basel) (2012)

Sketch of the equipment used for data collection. The camera captures a video stream for later inspection of behaviour. Both audio and video data are stored on an external hard drive. The dashed lines indicate microphone range and camera field of view.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3376600&req=5

f2-sensors-12-03773: Sketch of the equipment used for data collection. The camera captures a video stream for later inspection of behaviour. Both audio and video data are stored on an external hard drive. The dashed lines indicate microphone range and camera field of view.
Mentions: A combination of a shielded shotgun microphone (Sennheiser MKE 400) and a machine vision camera (uEye UI-1245LE-C) with a field of view (FOV) of 45° connected to a laptop were used for recordings. A multiple-shielded audio extension cable was used to minimise loss in fidelity. The camera and laptop were placed in a box at the edge of the field, whereas the microphone was placed 10 m in front of the camera, closer to the geese. The system was powered by two 12 V 92 Ah deep cycling car batteries and data were stored on 3 TB external hard drive. An overview is seen in Figure 2 (a detailed description can be found in Steen et al. [30]).

Bottom Line: A wide range of devices to detect and deter animals causing conflict are used for this purpose, although their effectiveness is often highly variable, due to habituation to disruptive or disturbing stimuli.The Support Vector Machine has proven to be a robust classifier for this kind of classification, as generality and non-linear capabilities are important.We conclude that vocalisations can be used to automatically detect behaviour of conflict wildlife species, and as such, may be used as an integrated part of a wildlife management system.

View Article: PubMed Central - PubMed

Affiliation: Department of Engineering, Aarhus University, Aarhus N, Denmark. kima.steen@agrsci.dk

ABSTRACT
Since human-wildlife conflicts are increasing, the development of cost-effective methods for reducing damage or conflict levels is important in wildlife management. A wide range of devices to detect and deter animals causing conflict are used for this purpose, although their effectiveness is often highly variable, due to habituation to disruptive or disturbing stimuli. Automated recognition of behaviours could form a critical component of a system capable of altering the disruptive stimuli to avoid this. In this paper we present a novel method to automatically recognise goose behaviour based on vocalisations from flocks of free-living barnacle geese (Branta leucopsis). The geese were observed and recorded in a natural environment, using a shielded shotgun microphone. The classification used Support Vector Machines (SVMs), which had been trained with labeled data. Greenwood Function Cepstral Coefficients (GFCC) were used as features for the pattern recognition algorithm, as they can be adjusted to the hearing capabilities of different species. Three behaviours are classified based in this approach, and the method achieves a good recognition of foraging behaviour (86-97% sensitivity, 89-98% precision) and a reasonable recognition of flushing (79-86%, 66-80%) and landing behaviour(73-91%, 79-92%). The Support Vector Machine has proven to be a robust classifier for this kind of classification, as generality and non-linear capabilities are important. We conclude that vocalisations can be used to automatically detect behaviour of conflict wildlife species, and as such, may be used as an integrated part of a wildlife management system.

No MeSH data available.


Related in: MedlinePlus