Sweet potato leaf detection in a natural scene based on faster R-CNN with a visual attention mechanism and DIoU-NMS

doi:10.1016/j.ecoinf.2022.101931

	Sweet potato leaf detection in a natural scene based on faster R-CNN with a visual attention mechanism and DIoU-NMS
	Wang, Mengxia 1,2; Fu, Boya 2; Fan, Jianbo 3; Wang, Yi 4; Zhang, Liankuan 5; Xia, Chunlei2
发表期刊	ECOLOGICAL INFORMATICS
ISSN	1574-9541
	2023-03-01
卷号	73 页码:12
关键词	Dense leaf detection Object detection CBAM Occlusion detection Plant leaves Leaf counting
DOI	10.1016/j.ecoinf.2022.101931
通讯作者	Xia, Chunlei([email protected])
英文摘要	Accurate detection of plant leaves is a meaningful and challenging task for developing smart agricultural sys-tems. To improve the performance of detecting plant leaves in natural scenes containing severe occlusion, overlapping, or shape variation, we developed an in situ sweet potato leaf detection method based on a modified Faster R-CNN framework and visual attention mechanism. First, a convolutional block attention module was added to the backbone network to enhance and extract critical features of leaf images by fusing cross-channel information and spatial information. Subsequently, the DIoU-NMS algorithm was adopted to modify the regional proposal network by replacing the original NMS. DIoU-NMS was utilized to reduce missed and incorrect detection in scenes of densely distributed leaves by considering the targets' overlap ratio, distance, and scale. The proposed leaf detection method was tested and evaluated on sweet potato plant images collected in agricultural fields. In the datasets, sweet potato leaves were presented in various sizes and poses, and a large proportion of leaves were occluded or overlapped with each other. The experimental results showed that the proposed leaf detection method outperforms state-of-the-art object detection methods. The mean average precision of the proposed method reached 95.7%, which was 2.9% higher than that of the original Faster R-CNN and 7.0% higher than that of YOLOv5. The proposed method achieved promising performance in detecting dense leaves or occluded leaves and could provide key techniques for applications in smart agriculture and ecological moni-toring, such as growth monitoring or plant phenotyping.
资助机构	National Key Research and Develop-ment Program of China ; Shandong Province Key R&D Program (Major Science and Technology Innovation Project)
收录类别	SCI
语种	英语
关键词[WOS]	IDENTIFICATION
研究领域[WOS]	Environmental Sciences & Ecology
WOS记录号	WOS:000895767300007
引用统计	被引频次：22[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.yic.ac.cn/handle/133337/36975
专题	中国科学院海岸带环境过程与生态修复重点实验室中国科学院海岸带环境过程与生态修复重点实验室_海岸带环境工程技术研究与发展中心
通讯作者	Xia, Chunlei
作者单位	1.Yantai Univ, Dept Phys & Elect Informat, Yantai 264005, Peoples R China 2.Chinese Acad Sci, Yantai Inst Coastal Zone Res, Yantai 264003, Peoples R China 3.Yantai Ctr Serv Conf & Exhibiting Ind, Yantai 264003, Peoples R China 4.Naval Aeronaut Univ, Coastal Def Coll, Yantai 264003, Peoples R China 5.South China Agr Univ, Coll Math & Informat, Guangzhou 510642, Peoples R China
推荐引用方式 GB/T 7714	Wang, Mengxia,Fu, Boya,Fan, Jianbo,et al. Sweet potato leaf detection in a natural scene based on faster R-CNN with a visual attention mechanism and DIoU-NMS[J]. ECOLOGICAL INFORMATICS,2023,73:12.
APA	Wang, Mengxia,Fu, Boya,Fan, Jianbo,Wang, Yi,Zhang, Liankuan,&Xia, Chunlei.(2023).Sweet potato leaf detection in a natural scene based on faster R-CNN with a visual attention mechanism and DIoU-NMS.ECOLOGICAL INFORMATICS,73,12.
MLA	Wang, Mengxia,et al."Sweet potato leaf detection in a natural scene based on faster R-CNN with a visual attention mechanism and DIoU-NMS".ECOLOGICAL INFORMATICS 73(2023):12.