LAUSR

Abstract Automatic Bird Species Recognition System helps ornithologists and researchers to study particular bird species, effect of climate changes, count of endangered species and their survival. Earlier researchers implemented this automation system using traditional methods such as Gaussian Mixture Model (GMM), Hidden Markov Model (HMM) and Dynamic Time Wrapping (DTW) etc. The efficiency with the systems mentioned above has shown very low accuracy and is time-consuming. The recognition system performance may be improved by using Spiking Neural Network (SNN), a third-generation artificial neural network (ANN). The main focus of work in this paper is to analyze sound waves produced by bird’s species. This work is based on the Attribute extrication methods (AEM). The paper briefly explains these methods and evaluates their performance on Spiking Neural Network classification. Spiking Neural Network classification using Permutation Pair Frequency Matrix (PPFM) proves to be a more efficient method in terms of accuracy percentage and lower computation time.