Identifying Leaf-Scale Wheat Aphids Using the Near-Ground Hyperspectral Pushbroom Imaging Spectrometer

Abstract : This study is to identify leaf-scale wheat aphids using the near-ground hyperspectral Pushbroom Imaging Spectrometer (PIS). Firstly, the spectral characteristics between normal and aphid-infested wheat leaves were compared in spectral reflectance. Concerning the serious aphid damage level, it is obvious that its spectral curve is badly flattened such as green peak (centered around 550 nm), red valley (centered around 680 nm), due to the influence of aphid. Specifically, in the visible spectrum (500-701 nm), the maximum delta (the maximum value minus the minimum value) is 3.3 and it is 7.5 in the near-infrared spectrum (701-900 nm). Then, the spectral difference and change rate were further analyzed. It seems that both curves show the mirror symmetry and their maximum values are 55.8% and 17.4%, respectively. For the difference curve, the value is negative in the visible spectrum (400-700 nm), which shows that the reflectance of normal wheat leaf is less than that of the serious level. Conversely, it is greater in the near-infrared spectrum (700-900 nm). Finally, based on the high spatial resolution PIS image, ENvironment for Visualizing Images (ENVI-EX) was utilized to extract aphids and the overall accuracy reaches 97%. The result indicates that the PIS is sufficient to identify the wheat aphids and this study can lay a foundation for further applications in precision agriculture using such a hyperspectral imaging system.
Type de document :
Communication dans un congrès
Daoliang Li; Yingyi Chen. 5th Computer and Computing Technologies in Agriculture (CCTA), Oct 2011, Beijing, China. Springer, IFIP Advances in Information and Communication Technology, AICT-369, pp.275-282, 2012, Computer and Computing Technologies in Agriculture V. 〈10.1007/978-3-642-27278-3_29〉
Liste complète des métadonnées

Littérature citée [16 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01360989
Contributeur : Hal Ifip <>
Soumis le : mardi 6 septembre 2016 - 15:10:43
Dernière modification le : mardi 6 septembre 2016 - 16:06:07
Document(s) archivé(s) le : mercredi 7 décembre 2016 - 14:20:32

Fichier

978-3-642-27278-3_29_Chapter.p...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Jinling Zhao, Dongyan Zhang, Juhua Luo, Dacheng Wang, Wenjiang Huang. Identifying Leaf-Scale Wheat Aphids Using the Near-Ground Hyperspectral Pushbroom Imaging Spectrometer. Daoliang Li; Yingyi Chen. 5th Computer and Computing Technologies in Agriculture (CCTA), Oct 2011, Beijing, China. Springer, IFIP Advances in Information and Communication Technology, AICT-369, pp.275-282, 2012, Computer and Computing Technologies in Agriculture V. 〈10.1007/978-3-642-27278-3_29〉. 〈hal-01360989〉

Partager

Métriques

Consultations de la notice

106

Téléchargements de fichiers

53