Open set recognition algorithm based on Conditional Gaussian Encoder

For the existing Closed Set Recognition (CSR) methods mistakenly identify unknown jamming signals as a known class, a Conditional Gaussian Encoder (CG-Encoder) for 1-dimensional signal Open Set Recognition (OSR) is designed. The network retains the original form of the signal as much as possible and...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yan Tang, Zhijin Zhao, Chun Li, Xueyi Ye
Formato:	article
Lenguaje:	EN
Publicado:	AIMS Press 2021
Materias:	jamming recognition open set recognition (osr) conditional gaussian encoder (cg-encoder) residual network kullback-leibler (kl) divergence Biotechnology TP248.13-248.65 Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/0db59c50f2dc4138b8690ef160f795d6
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:0db59c50f2dc4138b8690ef160f795d6
record_format	dspace
spelling	oai:doaj.org-article:0db59c50f2dc4138b8690ef160f795d62021-11-11T02:00:15ZOpen set recognition algorithm based on Conditional Gaussian Encoder10.3934/mbe.20213281551-0018https://doaj.org/article/0db59c50f2dc4138b8690ef160f795d62021-08-01T00:00:00Zhttps://www.aimspress.com/article/doi/10.3934/mbe.2021328?viewType=HTMLhttps://doaj.org/toc/1551-0018For the existing Closed Set Recognition (CSR) methods mistakenly identify unknown jamming signals as a known class, a Conditional Gaussian Encoder (CG-Encoder) for 1-dimensional signal Open Set Recognition (OSR) is designed. The network retains the original form of the signal as much as possible and deep neural network is used to extract useful information. CG-Encoder adopts residual network structure and a new Kullback-Leibler (KL) divergence is defined. In the training phase, the known classes are approximated to different Gaussian distributions in the latent space and the discrimination between classes is increased to improve the recognition performance of the known classes. In the testing phase, a specific and effective OSR algorithm flow is designed. Simulation experiments are carried out on 9 jamming types. The results show that the CSR and OSR performance of CG-Encoder is better than that of the other three kinds of network structures. When the openness is the maximum, the open set average accuracy of CG-Encoder is more than 70%, which is about 30% higher than the worst algorithm, and about 20% higher than the better one. When the openness is the minimum, the average accuracy of OSR is more than 95%.Yan TangZhijin ZhaoChun LiXueyi YeAIMS Pressarticlejamming recognitionopen set recognition (osr)conditional gaussian encoder (cg-encoder)residual networkkullback-leibler (kl) divergenceBiotechnologyTP248.13-248.65MathematicsQA1-939ENMathematical Biosciences and Engineering, Vol 18, Iss 5, Pp 6620-6637 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	jamming recognition open set recognition (osr) conditional gaussian encoder (cg-encoder) residual network kullback-leibler (kl) divergence Biotechnology TP248.13-248.65 Mathematics QA1-939
spellingShingle	jamming recognition open set recognition (osr) conditional gaussian encoder (cg-encoder) residual network kullback-leibler (kl) divergence Biotechnology TP248.13-248.65 Mathematics QA1-939 Yan Tang Zhijin Zhao Chun Li Xueyi Ye Open set recognition algorithm based on Conditional Gaussian Encoder
description	For the existing Closed Set Recognition (CSR) methods mistakenly identify unknown jamming signals as a known class, a Conditional Gaussian Encoder (CG-Encoder) for 1-dimensional signal Open Set Recognition (OSR) is designed. The network retains the original form of the signal as much as possible and deep neural network is used to extract useful information. CG-Encoder adopts residual network structure and a new Kullback-Leibler (KL) divergence is defined. In the training phase, the known classes are approximated to different Gaussian distributions in the latent space and the discrimination between classes is increased to improve the recognition performance of the known classes. In the testing phase, a specific and effective OSR algorithm flow is designed. Simulation experiments are carried out on 9 jamming types. The results show that the CSR and OSR performance of CG-Encoder is better than that of the other three kinds of network structures. When the openness is the maximum, the open set average accuracy of CG-Encoder is more than 70%, which is about 30% higher than the worst algorithm, and about 20% higher than the better one. When the openness is the minimum, the average accuracy of OSR is more than 95%.
format	article
author	Yan Tang Zhijin Zhao Chun Li Xueyi Ye
author_facet	Yan Tang Zhijin Zhao Chun Li Xueyi Ye
author_sort	Yan Tang
title	Open set recognition algorithm based on Conditional Gaussian Encoder
title_short	Open set recognition algorithm based on Conditional Gaussian Encoder
title_full	Open set recognition algorithm based on Conditional Gaussian Encoder
title_fullStr	Open set recognition algorithm based on Conditional Gaussian Encoder
title_full_unstemmed	Open set recognition algorithm based on Conditional Gaussian Encoder
title_sort	open set recognition algorithm based on conditional gaussian encoder
publisher	AIMS Press
publishDate	2021
url	https://doaj.org/article/0db59c50f2dc4138b8690ef160f795d6
work_keys_str_mv	AT yantang opensetrecognitionalgorithmbasedonconditionalgaussianencoder AT zhijinzhao opensetrecognitionalgorithmbasedonconditionalgaussianencoder AT chunli opensetrecognitionalgorithmbasedonconditionalgaussianencoder AT xueyiye opensetrecognitionalgorithmbasedonconditionalgaussianencoder
_version_	1718439559659782144

Open set recognition algorithm based on Conditional Gaussian Encoder

Ejemplares similares