A wheat spike detection method based on Transformer

doi:10.3389/fpls.2022.1023924

HFCAS OpenIR

	A wheat spike detection method based on Transformer
	Zhou, Qiong 1,2,3; Huang, Ziliang 1,2; Zheng, Shijian 1,4; Jiao, Lin 1,5; Wang, Liusan 1; Wang, Rujing1,2
	2022-10-20
发表期刊	FRONTIERS IN PLANT SCIENCE
ISSN	1664-462X
通讯作者	Jiao, Lin(ljiao@ahu.edu.cn) ; Wang, Liusan(lswang@iim.ac.cn) ; Wang, Rujing(rjwang@iim.ac.cn)
摘要	Wheat spike detection has important research significance for production estimation and crop field management. With the development of deep learning-based algorithms, researchers tend to solve the detection task by convolutional neural networks (CNNs). However, traditional CNNs equip with the inductive bias of locality and scale-invariance, which makes it hard to extract global and long-range dependency. In this paper, we propose a Transformer-based network named Multi-Window Swin Transformer (MW-Swin Transformer). Technically, MW-Swin Transformer introduces the ability of feature pyramid network to extract multi-scale features and inherits the characteristic of Swin Transformer that performs self-attention mechanism by window strategy. Moreover, bounding box regression is a crucial step in detection. We propose a Wheat Intersection over Union loss by incorporating the Euclidean distance, area overlapping, and aspect ratio, thereby leading to better detection accuracy. We merge the proposed network and regression loss into a popular detection architecture, fully convolutional one-stage object detection, and name the unified model WheatFormer. Finally, we construct a wheat spike detection dataset (WSD-2022) to evaluate the performance of the proposed methods. The experimental results show that the proposed network outperforms those state-of-the-art algorithms with 0.459 mAP (mean average precision) and 0.918 AP(50). It has been proved that our Transformer-based method is effective to handle wheat spike detection under complex field conditions.
关键词	deep learning IoU loss function transformer wheat spike detection agriculture
DOI	10.3389/fpls.2022.1023924
关键词[WOS]	DENSITY
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China ; Natural Science Foundation of Anhui Province ; [2019YFE0125700] ; [2208085MC57]
项目资助者	National Key R&D Program of China ; Natural Science Foundation of Anhui Province
WOS研究方向	Plant Sciences
WOS类目	Plant Sciences
WOS记录号	WOS:000879007100001
出版者	FRONTIERS MEDIA SA
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.hfcas.ac.cn:8080/handle/334002/130123
专题	中国科学院合肥物质科学研究院
通讯作者	Jiao, Lin; Wang, Liusan; Wang, Rujing
作者单位	1.Chinese Acad Sci, Inst Intelligent Machines, Hefei Inst Phys Sci, Hefei, Peoples R China 2.Univ Sci & Technol China, Sci Isl Branch, Hefei, Peoples R China 3.Anhui Agr Univ, Coll Informat & Comp, Hefei, Peoples R China 4.Univ Sci & Technol, Dept Informat Engn Southwest, Mianyang, Peoples R China 5.Anhui Univ, Sch Internet, Hefei, Peoples R China
推荐引用方式 GB/T 7714	Zhou, Qiong,Huang, Ziliang,Zheng, Shijian,et al. A wheat spike detection method based on Transformer[J]. FRONTIERS IN PLANT SCIENCE,2022,13.
APA	Zhou, Qiong,Huang, Ziliang,Zheng, Shijian,Jiao, Lin,Wang, Liusan,&Wang, Rujing.(2022).A wheat spike detection method based on Transformer.FRONTIERS IN PLANT SCIENCE,13.
MLA	Zhou, Qiong,et al."A wheat spike detection method based on Transformer".FRONTIERS IN PLANT SCIENCE 13(2022).