Hyperspectral Imaging for Detection of Leguminivora Glycinivorella Based on 3D Few-Shot Meta-Learning Model
GUI Jiang-sheng1, FEI Jing-yi1, FU Xia-ping2
1. School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China
2. Faculty of Mechanical Engineering & Automation, Zhejiang Sci-Tech University, Hangzhou 310018, China
Abstract:In order to reduce the influence of leguminivora glycinivorella on soybean production and quality, and to realize the rapid detection of leguminivora glycinivorella, this paper proposed a leguminivora glycinivorella detection model based on 3D-Realtion Network (3D-RN) model. Firstly, collect the hyperspectral images of 20 soybeans that are attached to eggs, larvae, gnawed and normal soybeans, respectively, and extract the region of interest (ROI) to establish a 3D-RN model based on hyperspectral images. The accuracy of the final model reached 82%±2.50%. Compared to the Model-Agnostic Meta-Learning (MAML) and Matching Network (MN) models, the 3D-RN model can fully measure the distance between sample features, and the recognition effect is greatly improved. Thus, this research shows that the 3D-RN model based on the hyperspectral image can detect leguminivora glycinivorella in a small number of samples. The method of combining few-shot meta-learning with hyperspectral provides a new idea for pest detection.
桂江生,费婧怡,傅霞萍. 三维小样本元学习模型的大豆食心虫虫害高光谱检测[J]. 光谱学与光谱分析, 2021, 41(07): 2171-2174.
GUI Jiang-sheng, FEI Jing-yi, FU Xia-ping. Hyperspectral Imaging for Detection of Leguminivora Glycinivorella Based on 3D Few-Shot Meta-Learning Model. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(07): 2171-2174.
[1] LI Zhao-jun, TIAN Ru-mei, PU Yan-yan, et al(李照君,田汝美,蒲艳艳,等). Soybean Science(大豆科学), 2020, 39(4): 577.
[2] GAO Yu, SHI Shu-sen(高 宇,史树森). Soybean Science (大豆科学), 2016, 35(6): 1025.
[3] DAI Yu-ting, ZHOU Bo, WANG Jun(代雨婷, 周 博, 王 俊). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2020, 36(3): 313.
[4] Tanaka F, Magariyama Y, Miyanoshita A. Food Chemistry, 2020, 303, 125381.
[5] Cui S, Ling P, Zhu H, et al. Sensors, 2018, 18(2): 378.
[6] Nanni L, Maguolo G, Pancino F. Ecological Informatics, 2020, 57: 101089.
[7] LUO Qing-qing, HUANG Tie-cheng, CHEN Shu-jiang, et al(罗青青, 黄铁成, 陈蜀江, 等). Jiangsu Journal of Agricultural Sciences (江苏农业学报), 2019, 35(4): 798.
[8] Li W, Chen P, Wang B, et al. Scientific Reports, 2019, 9(1):1.
[9] Rustia D J, Lin C E, Chung J, et al. Journal of Asia-Pacific Entomology, 2020, 23(1):17.
[10] LI Cui-ling, JIANG Kai, MA Wei, et al(李翠玲,姜 凯,马 伟,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2018, 38(1): 253.
[11] Lin T L, Chang H Y, Chen K H. Journal of Internet Technology, 2020, 21(2):605.
[12] Sung F, Yang Y, Zhang L, et al. Computer Vision and Pattern Recognition, 2018, 2018: 1199 (doi: 10.1109/CVPR.2018.00131).
[13] Finn C, Abbeel P, Levine S. Proc of Machine Loarning Research, 2017, 70: 1126.
[14] Vinyals O, Blundell C, Lillicrap T, et al. Matching Networks for One Shot Learning. Proceedings of the Neural Information Processing Systems Conference, 2016. 3630.