基于多尺度快速rcnn的飞机检测

Wei Miao, Z. Luo
{"title":"基于多尺度快速rcnn的飞机检测","authors":"Wei Miao, Z. Luo","doi":"10.1109/ICVRV.2018.00026","DOIUrl":null,"url":null,"abstract":"Remote sensing image recognition has been widely used in civil and military fieldsDIn view of plenty of interference factors in remote-sensing aircraftCsuch as shadeCnoiseCthe changing of perspectiveCetc. An improved target recognition algorithm in remote sensing image based on Faster-RCNN is proposed which uses a standard Region Proposal Network (RPN) generation and incorporates feature maps from shallower convolution feature maps. Convolution neural network is adopted to recognize aircraft target in complex environment , enhance the global context and local information to avoid information loss in the process of feature extractionCwhich improves recognition rate. Simulation results show that the feasibility of aircraft target recognition algorithm in remoting sensing image and the scale and posture changes of target can be overcome.MeanwhileCthe proposed algorithm has higher recognition effect and stronger robustness than traditional Faster-RCNN and BP neural network and support vector machine ( SVM) methods.","PeriodicalId":159517,"journal":{"name":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","volume":"50 2 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Aircraft Detection Based on Multiple Scale Faster-RCNN\",\"authors\":\"Wei Miao, Z. Luo\",\"doi\":\"10.1109/ICVRV.2018.00026\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Remote sensing image recognition has been widely used in civil and military fieldsDIn view of plenty of interference factors in remote-sensing aircraftCsuch as shadeCnoiseCthe changing of perspectiveCetc. An improved target recognition algorithm in remote sensing image based on Faster-RCNN is proposed which uses a standard Region Proposal Network (RPN) generation and incorporates feature maps from shallower convolution feature maps. Convolution neural network is adopted to recognize aircraft target in complex environment , enhance the global context and local information to avoid information loss in the process of feature extractionCwhich improves recognition rate. Simulation results show that the feasibility of aircraft target recognition algorithm in remoting sensing image and the scale and posture changes of target can be overcome.MeanwhileCthe proposed algorithm has higher recognition effect and stronger robustness than traditional Faster-RCNN and BP neural network and support vector machine ( SVM) methods.\",\"PeriodicalId\":159517,\"journal\":{\"name\":\"2018 International Conference on Virtual Reality and Visualization (ICVRV)\",\"volume\":\"50 2 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 International Conference on Virtual Reality and Visualization (ICVRV)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICVRV.2018.00026\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 International Conference on Virtual Reality and Visualization (ICVRV)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICVRV.2018.00026","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

摘要

由于遥感飞机存在阴影、噪声、视角变化等诸多干扰因素,遥感图像识别已广泛应用于民用和军事领域。提出了一种改进的基于Faster-RCNN的遥感图像目标识别算法,该算法使用标准的区域建议网络(RPN)生成,并结合浅卷积特征映射的特征映射。采用卷积神经网络对复杂环境下的飞机目标进行识别,增强了全局背景和局部信息,避免了特征提取过程中的信息丢失,提高了识别率。仿真结果表明,该算法在遥感图像中识别飞机目标是可行的,可以克服目标的尺度和姿态变化。与传统的Faster-RCNN、BP神经网络和支持向量机(SVM)方法相比,该算法具有更高的识别效果和更强的鲁棒性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Aircraft Detection Based on Multiple Scale Faster-RCNN
Remote sensing image recognition has been widely used in civil and military fieldsDIn view of plenty of interference factors in remote-sensing aircraftCsuch as shadeCnoiseCthe changing of perspectiveCetc. An improved target recognition algorithm in remote sensing image based on Faster-RCNN is proposed which uses a standard Region Proposal Network (RPN) generation and incorporates feature maps from shallower convolution feature maps. Convolution neural network is adopted to recognize aircraft target in complex environment , enhance the global context and local information to avoid information loss in the process of feature extractionCwhich improves recognition rate. Simulation results show that the feasibility of aircraft target recognition algorithm in remoting sensing image and the scale and posture changes of target can be overcome.MeanwhileCthe proposed algorithm has higher recognition effect and stronger robustness than traditional Faster-RCNN and BP neural network and support vector machine ( SVM) methods.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信