Transportation scene recognition based on high level feature representation

With the development of intelligent transportation,it has become an urgent problem to quickly and accurately recognize complex traffic scene.In recent years,a large number of scene recognition methods have been proposed to improve the effectiveness of traffic scene recognition,however,most of these...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenhua LIU, Yidong LI, Tao WANG, Jun WU, Yi JIN
Format: Article
Language:zho
Published: POSTS&TELECOM PRESS Co., LTD 2019-12-01
Series:智能科学与技术学报
Subjects:
Online Access:http://www.cjist.com.cn/thesisDetails#10.11959/j.issn.2096-6652.201943
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846171160686886912
author Wenhua LIU
Yidong LI
Tao WANG
Jun WU
Yi JIN
author_facet Wenhua LIU
Yidong LI
Tao WANG
Jun WU
Yi JIN
author_sort Wenhua LIU
collection DOAJ
description With the development of intelligent transportation,it has become an urgent problem to quickly and accurately recognize complex traffic scene.In recent years,a large number of scene recognition methods have been proposed to improve the effectiveness of traffic scene recognition,however,most of these algorithms cannot extract the semantic characteristics of the concept of vision,leading to the low recognition accuracy in traffic scenes.Therefore,a novel traffic scene recognition algorithm which extracts the high-level semantic and structural information for improving the accuracy was proposed.A system to discover semantically meaningful descriptions of the scene classes to reduce the “semantic gap” between the high level and the low-level feature representation was built.Then,the multi-label network was trained by minimizing loss function (namely,element-wise logistic loss) to obtain the high-level semantic representation of traffic scene images.Finally,experiments on four large-scale scene recognition datasets show that the proposed algorithm considerably outperforms other state-of-the-art methods.
format Article
id doaj-art-f8b37d622a734558af9affaf02f023a5
institution Kabale University
issn 2096-6652
language zho
publishDate 2019-12-01
publisher POSTS&TELECOM PRESS Co., LTD
record_format Article
series 智能科学与技术学报
spelling doaj-art-f8b37d622a734558af9affaf02f023a52024-11-11T06:51:29ZzhoPOSTS&TELECOM PRESS Co., LTD智能科学与技术学报2096-66522019-12-01139239959636794Transportation scene recognition based on high level feature representationWenhua LIUYidong LITao WANGJun WUYi JINWith the development of intelligent transportation,it has become an urgent problem to quickly and accurately recognize complex traffic scene.In recent years,a large number of scene recognition methods have been proposed to improve the effectiveness of traffic scene recognition,however,most of these algorithms cannot extract the semantic characteristics of the concept of vision,leading to the low recognition accuracy in traffic scenes.Therefore,a novel traffic scene recognition algorithm which extracts the high-level semantic and structural information for improving the accuracy was proposed.A system to discover semantically meaningful descriptions of the scene classes to reduce the “semantic gap” between the high level and the low-level feature representation was built.Then,the multi-label network was trained by minimizing loss function (namely,element-wise logistic loss) to obtain the high-level semantic representation of traffic scene images.Finally,experiments on four large-scale scene recognition datasets show that the proposed algorithm considerably outperforms other state-of-the-art methods.http://www.cjist.com.cn/thesisDetails#10.11959/j.issn.2096-6652.201943scene recognition;CNN;high-level feature;low-level feature
spellingShingle Wenhua LIU
Yidong LI
Tao WANG
Jun WU
Yi JIN
Transportation scene recognition based on high level feature representation
智能科学与技术学报
scene recognition;CNN;high-level feature;low-level feature
title Transportation scene recognition based on high level feature representation
title_full Transportation scene recognition based on high level feature representation
title_fullStr Transportation scene recognition based on high level feature representation
title_full_unstemmed Transportation scene recognition based on high level feature representation
title_short Transportation scene recognition based on high level feature representation
title_sort transportation scene recognition based on high level feature representation
topic scene recognition;CNN;high-level feature;low-level feature
url http://www.cjist.com.cn/thesisDetails#10.11959/j.issn.2096-6652.201943
work_keys_str_mv AT wenhualiu transportationscenerecognitionbasedonhighlevelfeaturerepresentation
AT yidongli transportationscenerecognitionbasedonhighlevelfeaturerepresentation
AT taowang transportationscenerecognitionbasedonhighlevelfeaturerepresentation
AT junwu transportationscenerecognitionbasedonhighlevelfeaturerepresentation
AT yijin transportationscenerecognitionbasedonhighlevelfeaturerepresentation