Multi-label UAV sound classification using Stacked Bidirectional LSTM
Dana Utebayeva, Akhan Almagambetov, Manal Alduraibi, Yelmurat Temirgaliyev, Lyazzat Ilipbayeva, Sunggat Marxuly
Abstract
Nowadays Unmanned Aerial Vehicles (UAVs) pose an increasing threat to public areas such as parks, schools, hospitals and official buildings. Different methods of dealing with UAV detection are developing more and more actively. This paper primarily focuses on two key aims: the first aim is to perform a multi-label classification system and the second aim is to develop Stacked Bidirectional Long Short-Term Memory (LSTM) with two hidden layers to categorize multiple UAVs sounds. Frame-wise spectral-domain features are applied as inputs of the proposed system. Overall, the results of the study show that the sound of UAVs can be classified into multiple labels. This study has been one of the first attempts to thoroughly examine Stacked Bidirectional LSTM for UAV sound classification task.