Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods
Rice diseases are major threats to rice yield and quality. Rapid and accurate detection of rice diseases is of great importance for precise disease prevention and treatment. Various spectroscopic techniques have been used to detect plant diseases. To rapidly and accurately detect three different ric...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2020-11-01
|
Series: | Frontiers in Plant Science |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2020.577063/full |
id |
doaj-2a7977417dde4961b1f8a68922251d06 |
---|---|
record_format |
Article |
spelling |
doaj-2a7977417dde4961b1f8a68922251d062020-11-25T04:09:10ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2020-11-011110.3389/fpls.2020.577063577063Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning MethodsLei Feng0Lei Feng1Baohua Wu2Baohua Wu3Susu Zhu4Susu Zhu5Junmin Wang6Zhenzhu Su7Fei Liu8Fei Liu9Yong He10Yong He11Chu Zhang12Chu Zhang13College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaInstitute of Crop Science and Nuclear Technology Utilization, Zhejiang Academy of Agricultural Sciences, Hangzhou, ChinaState Key Laboratory for Rice Biology, Institute of Biotechnology, Zhejiang University, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaKey Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, ChinaRice diseases are major threats to rice yield and quality. Rapid and accurate detection of rice diseases is of great importance for precise disease prevention and treatment. Various spectroscopic techniques have been used to detect plant diseases. To rapidly and accurately detect three different rice diseases [leaf blight (Xanthomonas oryzae pv. Oryzae), rice blast (Pyricularia oryzae), and rice sheath blight (Rhizoctonia solani)], three spectroscopic techniques were applied, including visible/near-infrared hyperspectral imaging (HSI) spectra, mid-infrared spectroscopy (MIR), and laser-induced breakdown spectroscopy (LIBS). Three different levels of data fusion (raw data fusion, feature fusion, and decision fusion) fusing three different types of spectral features were adopted to categorize the diseases of rice. Principal component analysis (PCA) and autoencoder (AE) were used to extract features. Identification models based on each technique and different fusion levels were built using support vector machine (SVM), logistic regression (LR), and convolution neural network (CNN) models. Models based on HSI performed better than those based on MIR and LIBS, with the accuracy over 93% for the test set based on PCA features of HSI spectra. The performance of rice disease identification varied with different levels of fusion. The results showed that feature fusion and decision fusion could enhance identification performance. The overall results illustrated that the three techniques could be used to identify rice diseases, and data fusion strategies have great potential to be used for rice disease detection.https://www.frontiersin.org/articles/10.3389/fpls.2020.577063/fullhyperspectral imagingmid-infrared spectroscopylaser-induced breakdown spectroscopydata fusionrice disease |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Lei Feng Lei Feng Baohua Wu Baohua Wu Susu Zhu Susu Zhu Junmin Wang Zhenzhu Su Fei Liu Fei Liu Yong He Yong He Chu Zhang Chu Zhang |
spellingShingle |
Lei Feng Lei Feng Baohua Wu Baohua Wu Susu Zhu Susu Zhu Junmin Wang Zhenzhu Su Fei Liu Fei Liu Yong He Yong He Chu Zhang Chu Zhang Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods Frontiers in Plant Science hyperspectral imaging mid-infrared spectroscopy laser-induced breakdown spectroscopy data fusion rice disease |
author_facet |
Lei Feng Lei Feng Baohua Wu Baohua Wu Susu Zhu Susu Zhu Junmin Wang Zhenzhu Su Fei Liu Fei Liu Yong He Yong He Chu Zhang Chu Zhang |
author_sort |
Lei Feng |
title |
Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods |
title_short |
Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods |
title_full |
Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods |
title_fullStr |
Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods |
title_full_unstemmed |
Investigation on Data Fusion of Multisource Spectral Data for Rice Leaf Diseases Identification Using Machine Learning Methods |
title_sort |
investigation on data fusion of multisource spectral data for rice leaf diseases identification using machine learning methods |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Plant Science |
issn |
1664-462X |
publishDate |
2020-11-01 |
description |
Rice diseases are major threats to rice yield and quality. Rapid and accurate detection of rice diseases is of great importance for precise disease prevention and treatment. Various spectroscopic techniques have been used to detect plant diseases. To rapidly and accurately detect three different rice diseases [leaf blight (Xanthomonas oryzae pv. Oryzae), rice blast (Pyricularia oryzae), and rice sheath blight (Rhizoctonia solani)], three spectroscopic techniques were applied, including visible/near-infrared hyperspectral imaging (HSI) spectra, mid-infrared spectroscopy (MIR), and laser-induced breakdown spectroscopy (LIBS). Three different levels of data fusion (raw data fusion, feature fusion, and decision fusion) fusing three different types of spectral features were adopted to categorize the diseases of rice. Principal component analysis (PCA) and autoencoder (AE) were used to extract features. Identification models based on each technique and different fusion levels were built using support vector machine (SVM), logistic regression (LR), and convolution neural network (CNN) models. Models based on HSI performed better than those based on MIR and LIBS, with the accuracy over 93% for the test set based on PCA features of HSI spectra. The performance of rice disease identification varied with different levels of fusion. The results showed that feature fusion and decision fusion could enhance identification performance. The overall results illustrated that the three techniques could be used to identify rice diseases, and data fusion strategies have great potential to be used for rice disease detection. |
topic |
hyperspectral imaging mid-infrared spectroscopy laser-induced breakdown spectroscopy data fusion rice disease |
url |
https://www.frontiersin.org/articles/10.3389/fpls.2020.577063/full |
work_keys_str_mv |
AT leifeng investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT leifeng investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT baohuawu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT baohuawu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT susuzhu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT susuzhu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT junminwang investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT zhenzhusu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT feiliu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT feiliu investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT yonghe investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT yonghe investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT chuzhang investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods AT chuzhang investigationondatafusionofmultisourcespectraldataforriceleafdiseasesidentificationusingmachinelearningmethods |
_version_ |
1724422925665173504 |