Audio-Based Drone Defect Detection Using Recurrence Plot and Deep Learning 


Vol. 48,  No. 1, pp. 114-122, Jan.  2023
10.7840/kics.2023.48.1.114


PDF
  Abstract

This paper proposes a drone defect detection scheme based on acoustic signals using Recurrence Plot (RP) and deep learning to detect drone defects and prevent damage caused by accidents in advance. The acoustic signals of a recurring pattern generated by the drone rotor are imaged using RP, and the drone defects are detected using the image recognition deep learning model YOLOv5. On the roof of a four-story building, RP datasets are constructed using the acoustic signals generated by the normal and abnormal drones composed of the Bebop 2 drone, and the YOLOv5 model is trained and inferred through the anaconda virtual environment on a Windows PC. Through comparative evaluation with mel-spectrogram, widely used for speech recognition, the classification accuracy and operation time are measured to prove the superiority of the performance of the proposed scheme. The accuracy of the proposed scheme is 97.5%, which is confirmed to be very accurate, and the future research direction is discussed.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

E. S. Kim and S. Y. Shin, "Audio-Based Drone Defect Detection Using Recurrence Plot and Deep Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 1, pp. 114-122, 2023. DOI: 10.7840/kics.2023.48.1.114.

[ACM Style]

Eun Seop Kim and Soo Young Shin. 2023. Audio-Based Drone Defect Detection Using Recurrence Plot and Deep Learning. The Journal of Korean Institute of Communications and Information Sciences, 48, 1, (2023), 114-122. DOI: 10.7840/kics.2023.48.1.114.

[KICS Style]

Eun Seop Kim and Soo Young Shin, "Audio-Based Drone Defect Detection Using Recurrence Plot and Deep Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 1, pp. 114-122, 1. 2023. (https://doi.org/10.7840/kics.2023.48.1.114)
Vol. 48, No. 1 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880