Passive acoustic detection and classification of marine mammal vocalizations

Acoustic methods are the best means for studying and monitoring marine mammals. Marine activities involve underwater sound emissions. Marine mammals are difficult to spot on the sea surface by visual observation, when weather and light conditions are poor. Many marine mammals produce loud and distinctive vocalizations, which can be detected more reliably than visual cues. Acoustic detection is dependent upon physical factors and species specific behavioral. Physical factor depends on physical acoustical properties such as source level, frequency range, and directionality of vocalizations. Behavioral factors refer to species and individual behavioral phenomenon that affect acoustical properties and signal detection ability. These may include vocalization rate, type of signal used, group size, depth and orientation of the animal while vocalizing. This paper uses ISHMAEL and PAMGAURD for passive acoustic detection, classification of marine mammal's vocalizations. Methods compared in this paper are energy sum, spectrogram correlation to perform detection and classification function using ISHMAEL and PAMGAURD. Results these methods are verified in MATLAB.