Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method

Object detection in remote sensing images plays an important role in both military and civilian remote sensing applications. Objects in remote sensing images are different from those in natural images. They have the characteristics of scale diversity, arbitrary directivity, and dense arrangement, wh...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Falin Wu, Jiaqi He, Guopeng Zhou, Haolun Li, Yushuang Liu, Xiaohong Sui
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	convolutional neural network (CNN) object detection remote sensing images three-point regression method (TPR) double detection head Science Q
Acceso en línea:	https://doaj.org/article/131d6f0c59f944118cbf710af380c109
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:131d6f0c59f944118cbf710af380c109
record_format	dspace
spelling	oai:doaj.org-article:131d6f0c59f944118cbf710af380c1092021-11-25T18:53:49ZImproved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method10.3390/rs132245172072-4292https://doaj.org/article/131d6f0c59f944118cbf710af380c1092021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4517https://doaj.org/toc/2072-4292Object detection in remote sensing images plays an important role in both military and civilian remote sensing applications. Objects in remote sensing images are different from those in natural images. They have the characteristics of scale diversity, arbitrary directivity, and dense arrangement, which causes difficulties in object detection. For objects with a large aspect ratio and that are oblique and densely arranged, using an oriented bounding box can help to avoid deleting some correct detection bounding boxes by mistake. The classic rotational region convolutional neural network (R2CNN) has advantages for text detection. However, R2CNN has poor performance in the detection of slender objects with arbitrary directivity in remote sensing images, and its fault tolerance rate is low. In order to solve this problem, this paper proposes an improved R2CNN based on a double detection head structure and a three-point regression method, namely, TPR-R2CNN. The proposed network modifies the original R2CNN network structure by applying a double fully connected (2-fc) detection head and classification fusion. One detection head is for classification and horizontal bounding box regression, the other is for classification and oriented bounding box regression. The three-point regression method (TPR) is proposed for oriented bounding box regression, which determines the positions of the oriented bounding box by regressing the coordinates of the center point and the first two vertices. The proposed network was validated on the DOTA-v1.5 and HRSC2016 datasets, and it achieved a mean average precision (mAP) of 3.90% and 15.27%, respectively, from feature pyramid network (FPN) baselines with a ResNet-50 backbone.Falin WuJiaqi HeGuopeng ZhouHaolun LiYushuang LiuXiaohong SuiMDPI AGarticleconvolutional neural network (CNN)object detectionremote sensing imagesthree-point regression method (TPR)double detection headScienceQENRemote Sensing, Vol 13, Iss 4517, p 4517 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	convolutional neural network (CNN) object detection remote sensing images three-point regression method (TPR) double detection head Science Q
spellingShingle	convolutional neural network (CNN) object detection remote sensing images three-point regression method (TPR) double detection head Science Q Falin Wu Jiaqi He Guopeng Zhou Haolun Li Yushuang Liu Xiaohong Sui Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
description	Object detection in remote sensing images plays an important role in both military and civilian remote sensing applications. Objects in remote sensing images are different from those in natural images. They have the characteristics of scale diversity, arbitrary directivity, and dense arrangement, which causes difficulties in object detection. For objects with a large aspect ratio and that are oblique and densely arranged, using an oriented bounding box can help to avoid deleting some correct detection bounding boxes by mistake. The classic rotational region convolutional neural network (R2CNN) has advantages for text detection. However, R2CNN has poor performance in the detection of slender objects with arbitrary directivity in remote sensing images, and its fault tolerance rate is low. In order to solve this problem, this paper proposes an improved R2CNN based on a double detection head structure and a three-point regression method, namely, TPR-R2CNN. The proposed network modifies the original R2CNN network structure by applying a double fully connected (2-fc) detection head and classification fusion. One detection head is for classification and horizontal bounding box regression, the other is for classification and oriented bounding box regression. The three-point regression method (TPR) is proposed for oriented bounding box regression, which determines the positions of the oriented bounding box by regressing the coordinates of the center point and the first two vertices. The proposed network was validated on the DOTA-v1.5 and HRSC2016 datasets, and it achieved a mean average precision (mAP) of 3.90% and 15.27%, respectively, from feature pyramid network (FPN) baselines with a ResNet-50 backbone.
format	article
author	Falin Wu Jiaqi He Guopeng Zhou Haolun Li Yushuang Liu Xiaohong Sui
author_facet	Falin Wu Jiaqi He Guopeng Zhou Haolun Li Yushuang Liu Xiaohong Sui
author_sort	Falin Wu
title	Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
title_short	Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
title_full	Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
title_fullStr	Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
title_full_unstemmed	Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method
title_sort	improved oriented object detection in remote sensing images based on a three-point regression method
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/131d6f0c59f944118cbf710af380c109
work_keys_str_mv	AT falinwu improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod AT jiaqihe improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod AT guopengzhou improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod AT haolunli improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod AT yushuangliu improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod AT xiaohongsui improvedorientedobjectdetectioninremotesensingimagesbasedonathreepointregressionmethod
_version_	1718410571793039360

Improved Oriented Object Detection in Remote Sensing Images Based on a Three-Point Regression Method

Ejemplares similares