USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS

Yung Hsu; Yen Chieh Ouyang; Jun Yi Lu; Mang Ou-Yang; Horng Yuh Guo; Tsang Sen Liu; Hsian Min Chen; Chao Cheng Wu; Chia Hsien Wen; Min Shao Shih; Chein I. Chang

doi:10.1109/IGARSS47720.2021.9555103

USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS

Yung Hsu, Yen Chieh Ouyang, Jun Yi Lu, Mang Ou-Yang, Horng Yuh Guo, Tsang Sen Liu, Hsian Min Chen, Chao Cheng Wu, Chia Hsien Wen, Min Shao Shih, Chein I. Chang

電機工程學系

研究成果: Paper › 同行評審

1 引文斯高帕斯（Scopus）

摘要

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).

原文	English
頁面	4416-4419
頁數	4
DOIs	https://doi.org/10.1109/IGARSS47720.2021.9555103
出版狀態	Published - 2021
事件	2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 - Brussels, Belgium 持續時間: 2021 7月 12 → 2021 7月 16

Conference

Conference	2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021
國家/地區	Belgium
城市	Brussels
期間	21-07-12 → 21-07-16

All Science Journal Classification (ASJC) codes

電腦科學應用
一般地球與行星科學

存取文件

10.1109/IGARSS47720.2021.9555103

其它文件與鏈接

引用此

Hsu, Y., Ouyang, Y. C., Lu, J. Y., Ou-Yang, M., Guo, H. Y., Liu, T. S., Chen, H. M., Wu, C. C., Wen, C. H., Shih, M. S., & Chang, C. I. (2021). USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS. 4416-4419. 論文發表於 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021, Brussels, Belgium. https://doi.org/10.1109/IGARSS47720.2021.9555103

@conference{66c07a8633ca4b2399a509a513d5258b,

title = "USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS",

abstract = "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).",

author = "Yung Hsu and Ouyang, {Yen Chieh} and Lu, {Jun Yi} and Mang Ou-Yang and Guo, {Horng Yuh} and Liu, {Tsang Sen} and Chen, {Hsian Min} and Wu, {Chao Cheng} and Wen, {Chia Hsien} and Shih, {Min Shao} and Chang, {Chein I.}",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 ; Conference date: 12-07-2021 Through 16-07-2021",

year = "2021",

doi = "10.1109/IGARSS47720.2021.9555103",

language = "English",

pages = "4416--4419",

}

Hsu, Y, Ouyang, YC, Lu, JY, Ou-Yang, M, Guo, HY, Liu, TS, Chen, HM, Wu, CC, Wen, CH, Shih, MS & Chang, CI 2021, 'USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS', 論文發表於 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021, Brussels, Belgium, 21-07-12 - 21-07-16 頁 4416-4419. https://doi.org/10.1109/IGARSS47720.2021.9555103

TY - CONF

T1 - USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS

AU - Hsu, Yung

AU - Ouyang, Yen Chieh

AU - Lu, Jun Yi

AU - Ou-Yang, Mang

AU - Guo, Horng Yuh

AU - Liu, Tsang Sen

AU - Chen, Hsian Min

AU - Wu, Chao Cheng

AU - Wen, Chia Hsien

AU - Shih, Min Shao

AU - Chang, Chein I.

PY - 2021

Y1 - 2021

N2 - 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).

AB - 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).

UR - http://www.scopus.com/inward/record.url?scp=85119326048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85119326048&partnerID=8YFLogxK

U2 - 10.1109/IGARSS47720.2021.9555103

DO - 10.1109/IGARSS47720.2021.9555103

M3 - Paper

AN - SCOPUS:85119326048

SP - 4416

EP - 4419

T2 - 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021

Y2 - 12 July 2021 through 16 July 2021

ER -

USING HYPERSPECTRAL IMAGING AND DEEP NEURAL NETWORK TO DETECT FUSARIUM WILT ON PHALAENOPSIS

摘要

Conference

All Science Journal Classification (ASJC) codes

存取文件

其它文件與鏈接

指紋

引用此