Detection of Bryde's whale short pulse calls using time domain features with hidden Markov models

Passive acoustic monitoring (PAM) is generally usedto extract signals produced by cetaceans. However, the large data volume from PAM process better analyzed using an automated technique such as hidden Markovmodels (HMM). In this paper, HMM used a detection and classification due to its robustness low time complexity. Nonetheless, certain parameters, choice of features be extracted signal, frame duration, number states affect performance model. Theresults show that exhibits best performances increases with short duration. increasing creates more computational complexity in The inshore Bryde's whales produce pulse calls distinct signal features, which are observable time-domain. Hence, time-domain feature vector utilized reduce HMM. Simulation results also average power provides compared other vectors for detecting call based on technique. More so, power, mean, zero-crossing rate, combined form single 3-dimensional (PaMZ). PaMZ-HMM shows improved reduced over existing extraction techniques Mel-scale frequency cepstral coefficients (MFCC) linear predictive coding (LPC). Thus, making suitable real-time detection.