TY - JOUR
T1 - Classification of animal dive tracks via automatic landmarking, principal components analysis and clustering
AU - Walker, Cameron
AU - MacKenzie, Monique Lea
AU - Donovan, Carl Robert
AU - Hastie, Gordon Drummond
AU - Quick, Nicola Jane
AU - Kidney, Darren
N1 - The BRS study was financially supported by the United States (U.S.) Office of Naval Research (www.onr.navy.mil) Grants N00014‐07‐10988, N00014‐07‐11023, N00014‐08‐10990; the U.S. Strategic Environmental Research and Development Program (www.serdp.org) Grant SI‐1539, the Environmental Readiness Division of the U.S. Navy (http://www.navy.mil/local/n45/), the U.S. Chief of Naval Operations Submarine Warfare Division (Undersea Surveillance), the U.S. National Oceanic and Atmospheric Administration (National Marine Fisheries Service, Office of Science and Technology) (http://www.st.nmfs.noaa.gov/), U.S. National Oceanic and Atmospheric Administration Ocean Acoustics Program (http://www.nmfs.noaa.gov/pr/acoustics/), and the Joint Industry Program on Sound and Marine Life of the International Association of Oil and Gas Producers (www.soundandmarinelife.org).
PY - 2011/8/19
Y1 - 2011/8/19
N2 - The behaviour of animals and their interactions with the environment can be inferred by tracking their movement. For this reason, biologgers are an important source of ecological data, but analysing the shape of the tracks they record is difficult. In this paper we present a technique for automatically determining landmarks that can be used to analyse the shape of animal tracks. The approach uses a parametric version of the SALSA algorithm to fit regression splines to 1‐dimensional curves in N dimensions (in practice N = 2 or 3). The knots of these splines are used as landmarks in a subsequent Principal Components Analysis, and the dives classified via agglomerative clustering. We demonstrate the efficacy of this algorithm on simulated 2‐dimensional dive data, and apply our method to real 3‐dimensional whale dive data from the Behavioral Response Study (BRS) in the Bahamas. The BRS is a series of experiments to quantify shifts in behavior due to SONAR. Our analysis of 3‐dimensional track data supports an alteration in the dive behavior post‐ensonification.
AB - The behaviour of animals and their interactions with the environment can be inferred by tracking their movement. For this reason, biologgers are an important source of ecological data, but analysing the shape of the tracks they record is difficult. In this paper we present a technique for automatically determining landmarks that can be used to analyse the shape of animal tracks. The approach uses a parametric version of the SALSA algorithm to fit regression splines to 1‐dimensional curves in N dimensions (in practice N = 2 or 3). The knots of these splines are used as landmarks in a subsequent Principal Components Analysis, and the dives classified via agglomerative clustering. We demonstrate the efficacy of this algorithm on simulated 2‐dimensional dive data, and apply our method to real 3‐dimensional whale dive data from the Behavioral Response Study (BRS) in the Bahamas. The BRS is a series of experiments to quantify shifts in behavior due to SONAR. Our analysis of 3‐dimensional track data supports an alteration in the dive behavior post‐ensonification.
KW - Automatic landmark generation
KW - Principal components analysis
KW - Regression spline
KW - Whale ensonification
U2 - 10.1890/ES11-00034.1
DO - 10.1890/ES11-00034.1
M3 - Article
SN - 2150-8925
VL - 2
SP - 1
EP - 13
JO - Ecosphere
JF - Ecosphere
IS - 8
ER -