Using Hyperspectral Imaging and Deep Neural Network to Detect Fusarium Wilton Phalaenopsis

In this paper, we combined hyperspectral imaging techniques and deep neural networks (DNN) to detect Fusarium wilt on Phalaenopsis. Spectral angle mapper (SAM) and constrained energy minimization (CEM) were used to find abnormal areas. Band selection (BS) methods include Harsanyi-Farrand-Chang (HFC)...

Full description

Saved in:
Bibliographic Details
Main Authors: Hsu, Yun, Ouyang, Yen-Chieh, Lu, Jun-Yi, Ou-Yang, Mang, Guo, Horng-Yuh, Liu, Tsang-Sen, Chen, Hsian-Min, Wu, Chao-Cheng, Wen, Chia-Hsien, Shih, Min-Shao, Chang, Chein-I
Format: Conference Proceeding
Language:English
Subjects:
Agricultural Product Inspection
Band Selection
Constrain Energy Minimization
Deep learning
Deep Neural Networks
Geoscience and remote sensing
HFC
Hybrid fiber coaxial cables
Hyperspectral Image
Imaging
Minimization
Neural networks
Spectral Angle Mapper
Support vector machines
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we combined hyperspectral imaging techniques and deep neural networks (DNN) to detect Fusarium wilt on Phalaenopsis. Spectral angle mapper (SAM) and constrained energy minimization (CEM) were used to find abnormal areas. Band selection (BS) methods include Harsanyi-Farrand-Chang (HFC), band priority (BP) and band decorrelation (BD) were applied to get effective bands. The results showed that, on the fifth day of Phalaenopsis infection, the best accuracy rates for detecting Fusarium wilt using VNIR and SWIR hyperspectral imaging were 93.5% and 94.9%, respectively. In most cases, the accuracy of using DNN is better than using support vector machine (SVM).
ISSN:2153-7003
DOI:10.1109/IGARSS47720.2021.9555103