Radar HRRP recognition based on CNN
In this study, ground target recognition based on one-dimensional convolutional neural network (CNN) is studied by exploiting the targets’ high-resolution range profiles (HRRPs). Contrary to conventional methods which need feature extraction artificially, CNN can automatically discover features for...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2019-09-01
|
Series: | The Journal of Engineering |
Subjects: | |
Online Access: | https://digital-library.theiet.org/content/journals/10.1049/joe.2019.0725 |
id |
doaj-5d455085b3ce4f2bb11ab6e654bccc62 |
---|---|
record_format |
Article |
spelling |
doaj-5d455085b3ce4f2bb11ab6e654bccc622021-04-02T11:12:27ZengWileyThe Journal of Engineering2051-33052019-09-0110.1049/joe.2019.0725JOE.2019.0725Radar HRRP recognition based on CNNJia Song0Yanhua Wang1Wei Chen2Yang Li3Junfu Wang4Beijing Institute of TechnologyBeijing Institute of TechnologyBeijing Institute of TechnologyBeijing Institute of TechnologyBeijing Racobit Electronic Information Technology Co., Ltd.In this study, ground target recognition based on one-dimensional convolutional neural network (CNN) is studied by exploiting the targets’ high-resolution range profiles (HRRPs). Contrary to conventional methods which need feature extraction artificially, CNN can automatically discover features for classification. The authors propose a multi-channel CNN architecture that can be applied on diverse forms of HRRP such as amplitude, complex, spectrum etc. Experimental results demonstrate the superiorities of the proposed method over conventional methods based on handcrafted features and single-channel CNN in terms of recognition accuracy. Visualisation of the ‘deep features’ shows higher separability than handcrafted features, thus providing an insight into its effectiveness in exploiting the intrinsic structures.https://digital-library.theiet.org/content/journals/10.1049/joe.2019.0725radar target recognitiontarget trackingradar resolutionfeature extractionradar computingconvolutional neural netsneural net architectureradar hrrp recognitionground target recognitionone-dimensional convolutional neural networkfeature extractionmultichannel cnn architecturesingle-channel cnndeep featureshigh-resolution range profiles |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jia Song Yanhua Wang Wei Chen Yang Li Junfu Wang |
spellingShingle |
Jia Song Yanhua Wang Wei Chen Yang Li Junfu Wang Radar HRRP recognition based on CNN The Journal of Engineering radar target recognition target tracking radar resolution feature extraction radar computing convolutional neural nets neural net architecture radar hrrp recognition ground target recognition one-dimensional convolutional neural network feature extraction multichannel cnn architecture single-channel cnn deep features high-resolution range profiles |
author_facet |
Jia Song Yanhua Wang Wei Chen Yang Li Junfu Wang |
author_sort |
Jia Song |
title |
Radar HRRP recognition based on CNN |
title_short |
Radar HRRP recognition based on CNN |
title_full |
Radar HRRP recognition based on CNN |
title_fullStr |
Radar HRRP recognition based on CNN |
title_full_unstemmed |
Radar HRRP recognition based on CNN |
title_sort |
radar hrrp recognition based on cnn |
publisher |
Wiley |
series |
The Journal of Engineering |
issn |
2051-3305 |
publishDate |
2019-09-01 |
description |
In this study, ground target recognition based on one-dimensional convolutional neural network (CNN) is studied by exploiting the targets’ high-resolution range profiles (HRRPs). Contrary to conventional methods which need feature extraction artificially, CNN can automatically discover features for classification. The authors propose a multi-channel CNN architecture that can be applied on diverse forms of HRRP such as amplitude, complex, spectrum etc. Experimental results demonstrate the superiorities of the proposed method over conventional methods based on handcrafted features and single-channel CNN in terms of recognition accuracy. Visualisation of the ‘deep features’ shows higher separability than handcrafted features, thus providing an insight into its effectiveness in exploiting the intrinsic structures. |
topic |
radar target recognition target tracking radar resolution feature extraction radar computing convolutional neural nets neural net architecture radar hrrp recognition ground target recognition one-dimensional convolutional neural network feature extraction multichannel cnn architecture single-channel cnn deep features high-resolution range profiles |
url |
https://digital-library.theiet.org/content/journals/10.1049/joe.2019.0725 |
work_keys_str_mv |
AT jiasong radarhrrprecognitionbasedoncnn AT yanhuawang radarhrrprecognitionbasedoncnn AT weichen radarhrrprecognitionbasedoncnn AT yangli radarhrrprecognitionbasedoncnn AT junfuwang radarhrrprecognitionbasedoncnn |
_version_ |
1724165448281358336 |