Multi-label UAV sound classification using Stacked Bidirectional LSTM

Dana Utebayeva; Akhan Almagambetov; Manal Alduraibi; Yelmurat Temirgaliyev; Lyazzat Ilipbayeva; Sungat Marxuly

doi:10.1109/IRC.2020.00086

Multi-label UAV sound classification using Stacked Bidirectional LSTM

Dana Utebayeva, Akhan Almagambetov, Manal Alduraibi, Yelmurat Temirgaliyev, Lyazzat Ilipbayeva, Sungat Marxuly

Department of Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

16 Citations (Scopus)

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.

Original language	English
Title of host publication	2020 Fourth IEEE International Conference on Robotic Computing (IRC)
Pages	453-458
Number of pages	6
DOIs	https://doi.org/10.1109/IRC.2020.00086
Publication status	Published - Nov 1 2020

Keywords

Computer architecture
Unmanned aerial vehicles
Noise measurement
Task analysis
Spectral analysis
Robots
Load modeling
sound classification
UAV sound classification
Multi label classification
LSTM
Stacked Bidirectional LSTM

Access to Document

10.1109/IRC.2020.00086

Cite this

@inproceedings{a7ef4f5348c7462197d06b83b811d2bd,

title = "Multi-label UAV sound classification using Stacked Bidirectional LSTM",

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.",

keywords = "Computer architecture, Unmanned aerial vehicles, Noise measurement, Task analysis, Spectral analysis, Robots, Load modeling, sound classification, UAV sound classification, Multi label classification, LSTM, Stacked Bidirectional LSTM",

author = "Dana Utebayeva and Akhan Almagambetov and Manal Alduraibi and Yelmurat Temirgaliyev and Lyazzat Ilipbayeva and Sungat Marxuly",

year = "2020",

month = nov,

day = "1",

doi = "10.1109/IRC.2020.00086",

language = "English",

pages = "453--458",

booktitle = "2020 Fourth IEEE International Conference on Robotic Computing (IRC)",

}

TY - GEN

T1 - Multi-label UAV sound classification using Stacked Bidirectional LSTM

AU - Utebayeva, Dana

AU - Almagambetov, Akhan

AU - Alduraibi, Manal

AU - Temirgaliyev, Yelmurat

AU - Ilipbayeva, Lyazzat

AU - Marxuly, Sungat

PY - 2020/11/1

Y1 - 2020/11/1

N2 - 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.

AB - 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.

KW - Computer architecture

KW - Unmanned aerial vehicles

KW - Noise measurement

KW - Task analysis

KW - Spectral analysis

KW - Robots

KW - Load modeling

KW - sound classification

KW - UAV sound classification

KW - Multi label classification

KW - LSTM

KW - Stacked Bidirectional LSTM

U2 - 10.1109/IRC.2020.00086

DO - 10.1109/IRC.2020.00086

M3 - Conference contribution

SP - 453

EP - 458

BT - 2020 Fourth IEEE International Conference on Robotic Computing (IRC)

ER -

Multi-label UAV sound classification using Stacked Bidirectional LSTM

Abstract

Keywords

Access to Document

Fingerprint

Cite this