Tags

Type your tag names separated by a space and hit enter

Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania pyloalis Larvae and Damage on Mulberry Leaves.
Sensors (Basel). 2018 Jun 28; 18(7)S

Abstract

Mulberry trees are an important crop for sericulture. Pests can affect the yield and quality of mulberry leaves. This study aims to develop a hyperspectral imaging system in visible and near-infrared (NIR) region (400⁻1700 nm) for the rapid identification of Diaphania pyloalis larvae and its damage. The extracted spectra of five region of interests (ROI), namely leaf vein, healthy mesophyll, slight damage, serious damage, and Diaphania pyloalis larva at 400⁻1000 nm (visible range) and 900⁻1700 nm (NIR range), were used to establish a partial least squares discriminant analysis (PLS-DA) and least-squares support vector machines (LS-SVM) models. Successive projections algorithm (SPA), uninformation variable elimination (UVE), UVE-SPA, and competitive adaptive reweighted sampling were used for variable selection. The best models in distinguishing between leaf vein, healthy mesophyll, slight damage and serious damage, leaf vein, healthy mesophyll, and larva, slight damage, serious damage, and larva were all the SPA-LS-SVM models, based on the NIR range data, and their correct rate of prediction (CRP) were all 100.00%. The best model for the identification of all five ROIs was the UVE-SPA-LS-SVM model, based on visible range data, which had the CRP value of 97.30%. In summary, visible and near infrared hyperspectral imaging could distinguish Diaphania pyloalis larvae and their damage from leaf vein and healthy mesophyll in a rapid and non-destructive way.

Authors+Show Affiliations

College of Animal Sciences, Zhejiang University, Hangzhou 310058, China. lxhuang@zju.edu.cn. South Taihu Agricultural Technology Extension Center in Huzhou, Zhejiang University, Huzhou 313000, China. lxhuang@zju.edu.cn.College of Animal Sciences, Zhejiang University, Hangzhou 310058, China. lyoung1101@163.com.College of Animal Sciences, Zhejiang University, Hangzhou 310058, China. 21517069@zju.edu.cn.College of Animal Sciences, Zhejiang University, Hangzhou 310058, China. m17794539235@163.com.College of Agriculture & Biotechnology, Zhejiang University, Zijingang Campus, Hangzhou 310058, China. 11216044@zju.edu.cn. Zhejiang Provincial Key Laboratory of Horticultural Plant Integrative Biology, Hangzhou 310058, China. 11216044@zju.edu.cn. The State Agriculture Ministry Laboratory of Horticultural Plant Growth, Development and Quality Improvement, Hangzhou 310058, China. 11216044@zju.edu.cn.Faculty of Mechanical Engineering & Automation, Zhejiang Sci-Tech University, Hangzhou 310018, China. fuxp@zstu.edu.cn.Faculty of Mechanical Engineering & Automation, Zhejiang Sci-Tech University, Hangzhou 310018, China. xqiangdu@zstu.edu.cn. Key Laboratory of Transplanting Equipment and Technology of Zhejiang Province, Hangzhou, 310018, China. xqiangdu@zstu.edu.cn.College of Agriculture & Biotechnology, Zhejiang University, Zijingang Campus, Hangzhou 310058, China. di_wu@zju.edu.cn. Zhejiang Provincial Key Laboratory of Horticultural Plant Integrative Biology, Hangzhou 310058, China. di_wu@zju.edu.cn. The State Agriculture Ministry Laboratory of Horticultural Plant Growth, Development and Quality Improvement, Hangzhou 310058, China. di_wu@zju.edu.cn.

Pub Type(s)

Journal Article

Language

eng

PubMed ID

29958467

Citation

Huang, Lingxia, et al. "Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania Pyloalis Larvae and Damage On Mulberry Leaves." Sensors (Basel, Switzerland), vol. 18, no. 7, 2018.
Huang L, Yang L, Meng L, et al. Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania pyloalis Larvae and Damage on Mulberry Leaves. Sensors (Basel). 2018;18(7).
Huang, L., Yang, L., Meng, L., Wang, J., Li, S., Fu, X., Du, X., & Wu, D. (2018). Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania pyloalis Larvae and Damage on Mulberry Leaves. Sensors (Basel, Switzerland), 18(7). https://doi.org/10.3390/s18072077
Huang L, et al. Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania Pyloalis Larvae and Damage On Mulberry Leaves. Sensors (Basel). 2018 Jun 28;18(7) PubMed PMID: 29958467.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Potential of Visible and Near-Infrared Hyperspectral Imaging for Detection of Diaphania pyloalis Larvae and Damage on Mulberry Leaves. AU - Huang,Lingxia, AU - Yang,Liang, AU - Meng,Liuwei, AU - Wang,Jingyu, AU - Li,Shaojia, AU - Fu,Xiaping, AU - Du,Xiaoqiang, AU - Wu,Di, Y1 - 2018/06/28/ PY - 2018/05/09/received PY - 2018/06/16/revised PY - 2018/06/26/accepted PY - 2018/7/1/entrez PY - 2018/7/1/pubmed PY - 2019/4/11/medline KW - Diaphania pyloalis KW - damage KW - hyperspectral imaging KW - larvae KW - mulberry leaves JF - Sensors (Basel, Switzerland) JO - Sensors (Basel) VL - 18 IS - 7 N2 - Mulberry trees are an important crop for sericulture. Pests can affect the yield and quality of mulberry leaves. This study aims to develop a hyperspectral imaging system in visible and near-infrared (NIR) region (400⁻1700 nm) for the rapid identification of Diaphania pyloalis larvae and its damage. The extracted spectra of five region of interests (ROI), namely leaf vein, healthy mesophyll, slight damage, serious damage, and Diaphania pyloalis larva at 400⁻1000 nm (visible range) and 900⁻1700 nm (NIR range), were used to establish a partial least squares discriminant analysis (PLS-DA) and least-squares support vector machines (LS-SVM) models. Successive projections algorithm (SPA), uninformation variable elimination (UVE), UVE-SPA, and competitive adaptive reweighted sampling were used for variable selection. The best models in distinguishing between leaf vein, healthy mesophyll, slight damage and serious damage, leaf vein, healthy mesophyll, and larva, slight damage, serious damage, and larva were all the SPA-LS-SVM models, based on the NIR range data, and their correct rate of prediction (CRP) were all 100.00%. The best model for the identification of all five ROIs was the UVE-SPA-LS-SVM model, based on visible range data, which had the CRP value of 97.30%. In summary, visible and near infrared hyperspectral imaging could distinguish Diaphania pyloalis larvae and their damage from leaf vein and healthy mesophyll in a rapid and non-destructive way. SN - 1424-8220 UR - https://www.unboundmedicine.com/medline/citation/29958467/Potential_of_Visible_and_Near_Infrared_Hyperspectral_Imaging_for_Detection_of_Diaphania_pyloalis_Larvae_and_Damage_on_Mulberry_Leaves_ L2 - https://www.mdpi.com/resolver?pii=s18072077 DB - PRIME DP - Unbound Medicine ER -