A model for predicting noise source-receiver distance based on an object detection function

Kang Dongshik 1 *, Iha Moritaka 1 and Thu Lan Nguyen 2

1 Department of Computer Science & Intelligent System, Faculty of Engineering, University of the Ryukyus, Japan
2 Department of Architecture and Production Design, Interdisciplinary Faculty of Science and Engineering, Shimane, University, Japan.
 
Research Article
International Journal of Science and Research Archive, 2024, 13(01), 541–548.
Article DOI: 10.30574/ijsra.2024.13.1.1667
Publication history: 
Received 31 July 2024; revised on 08 September 2024; accepted on 10 September 2024
 
Abstract: 
An accurate aircraft noise prediction model is necessary to predict the damage caused by expanded and new-constructed airport projects. Some widely-used models are constructed based on noise-power-distance (NPD) data appointed for each aircraft model. However, the lack of NPD data for some types of military aircraft makes it challenging to predict noise around airports that serve both civil and military purposes. Since NPD data were obtained based on manual measurement and often spends required much of human labor and dedicated measuring equipment, therefore, it is desirable to have an automatic system for this effort. This study proposes a model that estimates the distance from the recording point to the airplane, or noise source-receiver distance, based on the video that captures the flight movements around the airport and then provides reliable NPD data of a specific aircraft. In this study, the time-series data of the flight movement were separated into sound and image components. Then, the time-series images were analyzed and input to M2det for object detection. Finally, the noise source-receiver distance was estimated based on the length of the airplane in the image identified by the object detection function.
 
Keywords: 
Aircraft noise; Prediction model; Noise power distance (NPD); Object detection; M2det
 
Full text article in PDF: 
 
Copyright information: 

Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0