An Effective Data Augmentation Strategy for CNN-Based Pest Localization and Recognition in the Field

doi:10.1109/ACCESS.2019.2949852

CORC > 合肥物质科学研究院 > 中国科学院合肥物质科学研究院 > 中科院合肥智能机械研究所

	An Effective Data Augmentation Strategy for CNN-Based Pest Localization and Recognition in the Field
	Li, Rui 2,3,4; Wang, Rujing 2,3; Zhang, Jie 2,3; Xie, Chengjun 2,3; Liu, Liu 2,3,4; Wang, Fangyuan 2,3,4; Chen, Hongbo 2,3; Chen, Tianjiao 2,3; Hu, Haiying 2,3; Jia, Xiufang 2,3
刊名	IEEE ACCESS
	2019
卷号	7
关键词	Pest localization pest recognition convolutional neural network multi-scale data augmentation
ISSN号	2169-3536
DOI	10.1109/ACCESS.2019.2949852
通讯作者	Wang, Rujing(rjwang@iim.ac.cn) ; Zhang, Jie(76609080@qq.com) ; Xie, Chengjun(cjxie@iim.ac.cn)
英文摘要	In agriculture, pest always causes the major damage in fields and results in significant crop yield losses. Currently, manual pest classification and counting are very time-consuming and many subjective factors can affect the population counting accuracy. In addition, the existing pest localization and recognition methods based on Convolutional Neural Network (CNN) are not satisfactory for practical pest prevention in fields because of pests' different scales and attitudes. In order to address these problems, an effective data augmentation strategy for CNN-based method is proposed in this paper. In training phase, we adopt data augmentation through rotating images by various degrees followed by cropping into different grids. In this way, we could obtain a large number of extra multi-scale examples that could be adopted to train a multi-scale pest detection model. In terms of test phase, we utilize the test time augmentation (TTA) strategy that separately inferences input images with various resolutions using the trained multi-scale model. Finally, we fuse these detection results from different image scales by non-maximum suppression (NMS) for the final result. Experimental results on wheat sawfly, wheat aphid, wheat mite and rice planthopper in our domain specific dataset, show that our proposed data augmentation strategy achieves the pest detection performance of 81.4% mean Average Precision (mAP), which improves 11.63%, 7.93%,4.73% compared to three state-of-the-art approaches.
资助项目	National Key Technology Research and Development Program of China[2018YFD0200300] ; National Natural Science Foundation of China[31401293] ; National Natural Science Foundation of China[31671586] ; National Natural Science Foundation of China[61773360] ; Chinese Academy of Science and Technology Service Network Planning[KFJ-STS-ZDTP-048-02] ; Fundamental Research Funds for the Central Universities of China[ACAIM190101]
WOS关键词	IDENTIFICATION ; CLASSIFICATION ; NETWORKS ; INSECTS
WOS研究方向	Computer Science ; Engineering ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000498720000001
资助机构	National Key Technology Research and Development Program of China ; National Natural Science Foundation of China ; Chinese Academy of Science and Technology Service Network Planning ; Fundamental Research Funds for the Central Universities of China
内容类型	期刊论文
源URL	[http://ir.hfcas.ac.cn:8080/handle/334002/92789]
专题	合肥物质科学研究院_中科院合肥智能机械研究所
通讯作者	Wang, Rujing; Zhang, Jie; Xie, Chengjun
作者单位	1.Hefei Univ Technol, Sch Comp & Informat, Anhui Prov Key Lab Affect Comp & Adv Intelligent, Hefei 230009, Anhui, Peoples R China 2.Chinese Acad Sci, Inst Intelligent Machines, Hefei 230031, Anhui, Peoples R China 3.Chinese Acad Sci, Hefei Inst Phys Sci, Hefei 230031, Anhui, Peoples R China 4.Univ Sci & Technol China, Dept Automat, Hefei 230026, Anhui, Peoples R China 5.Natl Agrotech Extens & Serv Ctr, Beijing 100125, Peoples R China
推荐引用方式 GB/T 7714	Li, Rui,Wang, Rujing,Zhang, Jie,et al. An Effective Data Augmentation Strategy for CNN-Based Pest Localization and Recognition in the Field[J]. IEEE ACCESS,2019,7.
APA	Li, Rui.,Wang, Rujing.,Zhang, Jie.,Xie, Chengjun.,Liu, Liu.,...&Liu, Wancai.(2019).An Effective Data Augmentation Strategy for CNN-Based Pest Localization and Recognition in the Field.IEEE ACCESS,7.
MLA	Li, Rui,et al."An Effective Data Augmentation Strategy for CNN-Based Pest Localization and Recognition in the Field".IEEE ACCESS 7(2019).