PestNet: An End-to-End Deep Learning Approach for Large-Scale Multi-Class Pest Detection and Classification

被引:234
作者
Liu, Liu [1 ,2 ,3 ]
Wang, Rujing [1 ,2 ]
Xie, Chengjun [1 ,2 ]
Yang, Po [4 ]
Wang, Fangyuan [1 ,2 ,3 ]
Sudirman, Sud [4 ]
Liu, Wancai [5 ]
机构
[1] Chinese Acad Sci, Inst Intelligent Machines, Hefei 230031, Anhui, Peoples R China
[2] Chinese Acad Sci, Hefei Inst Phys Sci, Hefei 230031, Anhui, Peoples R China
[3] Univ Sci & Technol China, Dept Automat, Hefei 230026, Anhui, Peoples R China
[4] Liverpool John Moores Univ, Dept Comp Sci, Liverpool L3 3AF, Merseyside, England
[5] Natl Agrotech Extens & Serv Ctr, Beijing 100125, Peoples R China
基金
中国国家自然科学基金;
关键词
Channel-spatial attention; convolutional neural network; multi-class pest detection; position-sensitive score map; region proposal network; IDENTIFICATION; INSECTS;
D O I
10.1109/ACCESS.2019.2909522
中图分类号
TP [自动化技术、计算机技术];
学科分类号
080201 [机械制造及其自动化];
摘要
Multi-class pest detection is one of the crucial components in pest management involving localization in addition to classification which is much more difficult than generic object detection because of the apparent differences among pest species. This paper proposes a region-based end-to-end approach named PestNet for large-scale multi-class pest detection and classification based on deep learning. PestNet consists of three major parts. First, a novel module channel-spatial attention (CSA) is proposed to be fused into the convolutional neural network (CNN) backbone for feature extraction and enhancement. The second one is called region proposal network (RPN) that is adopted for providing region proposals as potential pest positions based on extracted feature maps from images. Position-sensitive score map (PSSM), the third component, is used to replace fully connected (FC) layers for pest classification and bounding box regression. Furthermore, we apply contextual regions of interest (RoIs) as contextual information of pest features to improve detection accuracy. We evaluate PestNet on our newly collected large-scale pests' image dataset, Multi-class Pests Dataset 2018 (MPD2018) captured by our designed task-specific image acquisition equipment, covering more than 80k images with over 580k pests labeled by agricultural experts and categorized in 16 classes. The experimental results show that the proposed PestNet performs well on multi-class pest detection with 75.46% mean average precision (mAP), which outperforms the state-of-the-art methods.
引用
收藏
页码:45301 / 45312
页数:12
相关论文
共 42 条
[1]
[Anonymous], PROC CVPR IEEE
[2]
[Anonymous], 2001, ORDINAL LOGISTIC REG
[3]
[Anonymous], PROC CVPR IEEE
[4]
[Anonymous], ADV NEURAL INFORM PR, DOI DOI 10.1109/TPAMI.2016.2577031
[5]
[Anonymous], 2016, ADV NEURAL INFORM PR
[6]
[Anonymous], COMPUT ELECTRON AGRI
[7]
[Anonymous], 2007, RED
[8]
Bengio Y., 2012, P ICML WORKSH UNS TR, V7, P19, DOI DOI 10.5555/3045796.3045800
[9]
Bottou Leon, 2012, Neural Networks: Tricks of the Trade. Second Edition: LNCS 7700, P421, DOI 10.1007/978-3-642-35289-8_25
[10]
Caruana R, 2001, ADV NEUR IN, V13, P402