Aircraft Noise Detection Based on SVM optimized with Genetic Algorithm
Ding Jianli -, Yang Yong · Journal of Convergence Information Technology · 2013
This paper presents a support vector machine classifier for aircraft noise recognition which exists as one of the most difficult tasks involved in the process of noise monitoring near airports. The support vector machine classifier presented in this paper is used to pick out all sound clips of aircraft noise from all noise files after feature-extraction step. The selection of SVM parameters has a great impact on its performance. However, there exist few comprehensive theories about the selection of the SVM parameters. In this paper, we optimize the SVM classifier with Genetic Algorithm by searching rather better parameters. Experiment result show that the detection accuracy of the whole algorithm can reach 90% or more in different degrees of noise environment.