A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression

Vehicle detection plays a vital role in the design of Automatic Driving System (ADS), which has achieved remarkable improvements in recent years. However, vehicle detection in night scenes still has considerable challenges for the reason that the vehicle features are not obvious and are easily affec...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yan Liu, Tiantian Qiu, Jingwen Wang, Wenting Qi
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	nighttime vehicle detection attentive GAN multiple local regression improved RoI pooling Science Q Astrophysics QB460-466 Physics QC1-999
Acceso en línea:	https://doaj.org/article/b54d27ff797a4edeb72f37e1ce7d72ae
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b54d27ff797a4edeb72f37e1ce7d72ae
record_format	dspace
spelling	oai:doaj.org-article:b54d27ff797a4edeb72f37e1ce7d72ae2021-11-25T17:30:06ZA Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression10.3390/e231114901099-4300https://doaj.org/article/b54d27ff797a4edeb72f37e1ce7d72ae2021-11-01T00:00:00Zhttps://www.mdpi.com/1099-4300/23/11/1490https://doaj.org/toc/1099-4300Vehicle detection plays a vital role in the design of Automatic Driving System (ADS), which has achieved remarkable improvements in recent years. However, vehicle detection in night scenes still has considerable challenges for the reason that the vehicle features are not obvious and are easily affected by complex road lighting or lights from vehicles. In this paper, a high-accuracy vehicle detection algorithm is proposed to detect vehicles in night scenes. Firstly, an improved Generative Adversarial Network (GAN), named Attentive GAN, is used to enhance the vehicle features of nighttime images. Then, with the purpose of achieving a higher detection accuracy, a multiple local regression is employed in the regression branch, which predicts multiple bounding box offsets. An improved Region of Interest (RoI) pooling method is used to get distinguishing features in a classification branch based on Faster Region-based Convolutional Neural Network (R-CNN). Cross entropy loss is introduced to improve the accuracy of classification branch. The proposed method is examined with the proposed dataset, which is composed of the selected nighttime images from BDD-100k dataset (Berkeley Diverse Driving Database, including 100,000 images). Compared with a series of state-of-the-art detectors, the experiments demonstrate that the proposed algorithm can effectively contribute to vehicle detection accuracy in nighttime.Yan LiuTiantian QiuJingwen WangWenting QiMDPI AGarticlenighttime vehicle detectionattentive GANmultiple local regressionimproved RoI poolingScienceQAstrophysicsQB460-466PhysicsQC1-999ENEntropy, Vol 23, Iss 1490, p 1490 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	nighttime vehicle detection attentive GAN multiple local regression improved RoI pooling Science Q Astrophysics QB460-466 Physics QC1-999
spellingShingle	nighttime vehicle detection attentive GAN multiple local regression improved RoI pooling Science Q Astrophysics QB460-466 Physics QC1-999 Yan Liu Tiantian Qiu Jingwen Wang Wenting Qi A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
description	Vehicle detection plays a vital role in the design of Automatic Driving System (ADS), which has achieved remarkable improvements in recent years. However, vehicle detection in night scenes still has considerable challenges for the reason that the vehicle features are not obvious and are easily affected by complex road lighting or lights from vehicles. In this paper, a high-accuracy vehicle detection algorithm is proposed to detect vehicles in night scenes. Firstly, an improved Generative Adversarial Network (GAN), named Attentive GAN, is used to enhance the vehicle features of nighttime images. Then, with the purpose of achieving a higher detection accuracy, a multiple local regression is employed in the regression branch, which predicts multiple bounding box offsets. An improved Region of Interest (RoI) pooling method is used to get distinguishing features in a classification branch based on Faster Region-based Convolutional Neural Network (R-CNN). Cross entropy loss is introduced to improve the accuracy of classification branch. The proposed method is examined with the proposed dataset, which is composed of the selected nighttime images from BDD-100k dataset (Berkeley Diverse Driving Database, including 100,000 images). Compared with a series of state-of-the-art detectors, the experiments demonstrate that the proposed algorithm can effectively contribute to vehicle detection accuracy in nighttime.
format	article
author	Yan Liu Tiantian Qiu Jingwen Wang Wenting Qi
author_facet	Yan Liu Tiantian Qiu Jingwen Wang Wenting Qi
author_sort	Yan Liu
title	A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
title_short	A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
title_full	A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
title_fullStr	A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
title_full_unstemmed	A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression
title_sort	nighttime vehicle detection method with attentive gan for accurate classification and regression
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/b54d27ff797a4edeb72f37e1ce7d72ae
work_keys_str_mv	AT yanliu anighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT tiantianqiu anighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT jingwenwang anighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT wentingqi anighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT yanliu nighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT tiantianqiu nighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT jingwenwang nighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression AT wentingqi nighttimevehicledetectionmethodwithattentiveganforaccurateclassificationandregression
_version_	1718412306496356352

A Nighttime Vehicle Detection Method with Attentive GAN for Accurate Classification and Regression

Ejemplares similares