A STUDY OF REAL-TIME RECOGNITION OF UNMANNED AERIAL VEHICLES IN OUTDOOR AREAS BASED ON A RANDOM FOREST ALGORITHM

IF 0.8 Q3 ENGINEERING, AEROSPACE

Aviation Pub Date : 2022-11-15 DOI:10.3846/aviation.2022.17910

Zhou-Tai Tian, Daojie Yu

{"title":"A STUDY OF REAL-TIME RECOGNITION OF UNMANNED AERIAL VEHICLES IN OUTDOOR AREAS BASED ON A RANDOM FOREST ALGORITHM","authors":"Zhou-Tai Tian, Daojie Yu","doi":"10.3846/aviation.2022.17910","DOIUrl":null,"url":null,"abstract":"With the widespread use of unmanned aerial vehicles (UAVs) in life, the real-time recognition of UAVs has become an important issue. The authors of this paper mainly studied the application of the random forest (RF) algorithm in the outdoor real-time recognition of UAVs. Mel-Frequency Cepstral Coefficient (MFCC) features were extracted from sound signals firstly, and then the RF method was combined with weighted voting to obtain the improved random forest (IRF) method to identify UAV sounds and environmental sounds. An experimental analysis was conducted. The modeling time of the IRF method increased by 9.52% compared with the RF method, and the recognition rate of the IRF method decreased with the increase of the distance from the microphone; however, the recognition rate of the IRF method was always higher than that of the RF method, and the recognition rate of the IRF method for the mixed samples was always higher than 90%. When the distance was 10 m, the IRF method still had a recognition rate of 91.29%. The experimental results verify the effectiveness of the IRF method for the outdoor real-time recognition of UAVs and its practical application feasibility.","PeriodicalId":51910,"journal":{"name":"Aviation","volume":" ","pages":""},"PeriodicalIF":0.8000,"publicationDate":"2022-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Aviation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3846/aviation.2022.17910","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, AEROSPACE","Score":null,"Total":0}

引用次数: 0

Abstract

With the widespread use of unmanned aerial vehicles (UAVs) in life, the real-time recognition of UAVs has become an important issue. The authors of this paper mainly studied the application of the random forest (RF) algorithm in the outdoor real-time recognition of UAVs. Mel-Frequency Cepstral Coefficient (MFCC) features were extracted from sound signals firstly, and then the RF method was combined with weighted voting to obtain the improved random forest (IRF) method to identify UAV sounds and environmental sounds. An experimental analysis was conducted. The modeling time of the IRF method increased by 9.52% compared with the RF method, and the recognition rate of the IRF method decreased with the increase of the distance from the microphone; however, the recognition rate of the IRF method was always higher than that of the RF method, and the recognition rate of the IRF method for the mixed samples was always higher than 90%. When the distance was 10 m, the IRF method still had a recognition rate of 91.29%. The experimental results verify the effectiveness of the IRF method for the outdoor real-time recognition of UAVs and its practical application feasibility.

查看原文本刊更多论文

基于随机森林算法的户外无人机实时识别研究

随着无人机在生活中的广泛应用，对无人机的实时识别已成为一个重要问题。本文主要研究了随机森林算法在无人机户外实时识别中的应用。首先从声音信号中提取Mel-Frequency倒谱系数(MFCC)特征，然后将射频方法与加权投票相结合，得到改进的随机森林(IRF)方法，用于识别无人机声音和环境声音。进行了实验分析。与射频方法相比，IRF方法的建模时间增加了9.52%，随着与传声器距离的增加，IRF方法的识别率降低;但IRF法的识别率始终高于RF法，且IRF法对混合样品的识别率始终高于90%。当距离为10 m时，IRF方法的识别率仍为91.29%。实验结果验证了IRF方法用于无人机室外实时识别的有效性和实际应用的可行性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Aviation ENGINEERING, AEROSPACE-

CiteScore

2.40

自引率

10.00%

发文量

审稿时长

15 weeks

期刊介绍： CONCERNING THE FOLLOWING FIELDS OF RESEARCH: ▪ Flight Physics ▪ Air Traffic Management ▪ Aerostructures ▪ Airports ▪ Propulsion ▪ Human Factors ▪ Aircraft Avionics, Systems and Equipment ▪ Air Transport Technologies and Development ▪ Flight Mechanics ▪ History of Aviation ▪ Integrated Design and Validation (method and tools) Besides, it publishes: short reports and notes, reviews, reports about conferences and workshops