Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/98134
標題: 使用高光譜波段選擇技術檢測蝴蝶蘭黃葉病
Detection of Fusarium Wilt on Phalaenopsis Using Hyperspectral Band Selection Techniques
作者: 李孟爵
Meng-Chueh Lee
關鍵字: 農產品品質控管
高光譜影像
黃葉病檢測
波段優先
波段選擇
能量最小化限制
光譜訊息離散度
Agricultural product inspection
Hyperspectral image
Fusarium Wilt on Phalaenopsis
Band prioritization(BP)
Band selection(BS)
Constrain energy minimization(CEM)
Spectral information divergence(SID)
SeQuential N-FINDR
引用: [1] C. I Chang, 'Hyperspectral Imaging: Techniques for Spectral Detection and Classification' Kluwer, 2003. [2] C. I Chang, 'Hyperspectral Data Processing: Signal Processing Algorithm Design and Analysis' Wiley, 2013. [3] C. C. Wu, Y.H. Liao, W.S. Lo, H.Y. Guo, C. Lin, C.H. Wen, H.M. Chen, Y.C. Ouyang, C. I. Chang. 'Band Weighting Spectral Measurement for Detection of Pesticide Residues using Hyperspectral Remote Sensing' IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Milan, Italy, 26-31 July, 2015. [4] D. Snadya. 'Post-processing and Band Selection for Hyperspectral Image Data Classification with AdaBoost.MH' IEEE International Conference on Sustainable Information Engineering and Technology (SIET), 2017. [5] N. Hu, D. Wei, L. Zhang, J. Wang, H. Xu, Y. Zhao. 'Application of Vis-NIR Hyperspectral Imaging in Agricultural Products Detection' IEEE International Conference on Advance Infocomm Technology (ICAIT), 9th, 2017. [6] H. Li, et al., 'Nondestructive detection of total volatile basic nitrogen (TVB-N) content in pork meat by integrating hyperspectral imaging and colorimetric sensor combined with a nonlinear data fusion.' LWT- Food Science and Technology, 2015. 63(1): p. 268-274. [7] C. I Chang, Q. Du, T. S. Sun, and M. L. G. Althouse, 'A joint band prioritization and band decorrelation approach to band selection for hyperspectral image classification' IEEE Transactions on Geoscience Remote Sensing, vol. 37, no. 6, pp. 2631–2641, Nov. 1999. [8] C. I Chang, and K. H. Liu, 'Progressive Band Selection of Spectral Unmixing for Hyperspectral Imagery' IEEE Transactions on Geoscience Remote Sensing, vol. 52, No. 4, April, 2014. [9] K. Y. Ma, Y.M. Kuo, Y.C. Ouyang, C. I Chang. 'Improving Pesticide Residues Detection Using Band Prioritization and Constrained Energy Minimization' IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Fort Worth, TX, USA, 23-28 July, 2017. [10] J. B. Lee, A. S. Woodyatt, and M. Berman, 'Enhancement of high spectral resolution remote sensing data by a noise-adjusted principal components transform' IEEE Transactions on Geoscience Remote Sensing, vol. 28, no. 3, pp. 295–304, May 1990. [11] Y. Xue, Q. Wang, Q. Li, M. Zhou, H. Liu, 'Blood Cell Segmentation Based on the Hybrid Algorithm of Spectral Angle Mapping and Spectral Information Divergence' IEEE International Congress on Image and Signal Processing, BioMedical Engineering and Informatics(CISP-BMEI), Datong, China, 15-17 Oct., 2016. [12] C. I Chang, 'Spectral Information Divergence for Hyperspectral Image Analysis' IEEE International Geoscience and Remote Sensing Symposium (IGARSS), vol. 1, 1999. [13] J. C. Harsanyi, 'Detection and classification of Subpixel Spectral Signatures in Hyperspectral Image Sequences' Ph.D. dissertation, Department of Electrical Engineering, University of Maryland, Baltimore County, MD, 1993. [14] C. I Chang, 'Target signature-constrained mixed pixel classification for hyperspectral imagery' IEEE Transactions on Geoscience Remote Sensing, vol. 40, no. 5, pp. 1065–1081, 2002. [15] D. U. Qian, R, Hsuan, C. I Chang, 'A comparative study for orthogonal subspace projection and constrained energy minimization.' IEEE Transactions on Geoscience Remote Sensing, vol. 41, no. 6, pp. 1525–1529, 2003. [16] C. I Chang, 'Hyperspectral Imaging: Techniques for Spectral Detection and Classification.' Norway, MA, USA: Kluwer/Plenum, 2003. [17] C. I Chang, 'A Review of Virtual Dimensionality for Hyperspectral Imagery.' IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol.11, no. 4, April, 2018.
摘要: 在蘭花的產業中,蝴蝶蘭深受消費者喜愛,台灣的蝴蝶蘭有超過九成出口,是全球蝴蝶蘭供應主力,也是國內精緻農業外銷金額首位,主要外銷的地點以美國以及日本佔最大宗。過去的研究報告顯示,黃葉病一直是花農所頭痛的疾病,這種病不至於使植株立即死亡,但卻非常相關的影響了蝴蝶蘭的品質。近年來,農產品的品質控管一直是大眾所重視的問題。然而現階段所使用的農產品品質管控方法,大多是採用人工抽樣檢測來判斷品質優劣,其實檢驗程序不但複雜,也無法得到即時的檢測結果,某些破壞性的檢測方式,也會造成受檢測的樣本破壞使其無法再銷售。高光譜影像(hyperspectral image)是一取得整個跨電磁波領域資訊的技術,透過光在不同目標物上所產生的反射光譜差異,可檢測出不同目標物的組成成分,此方法目前也被廣泛的應用於農產品品質檢測中。本論文主要介紹蝴蝶蘭黃葉病的偵測方法,首先先將樣本分成兩個部分,分別是健康組與感病組。之後利用波段選擇(Band Selection)技術萃取出蝴蝶蘭黃葉病的特徵光譜,其技術為使用不同的波段優先(Band Prioritization)統計量搭配波段去相關(Band De-correlation)去掉相關性較接近的波段。同時我們也利用一些演算法:限制能量最小化法(CEM)、光譜訊息離散度(SID)與SQ-FINDR來驗證挑選的波段是否為黃葉病的特徵波段。我們希望這個研究能幫助到農民,使其在外銷前就能得知植株是否有感染黃葉病的可能,以降低他們的損失。
Phalaenopsis is a significant agricultural product with high economic value in Taiwan and more than 90% of Phalaenopsis is exported to all over the world. In recent reports, Fusarium wilt on Phalaenopsis is a disease that makes farmers suffer seriously. It causes Phalaenopsis leaves to turn yellow, dwindle, dehydrate and finally die. Although Phalaenopsis does not die immediately with Fusarium wilt, it seriously decreases the quality that buyers cannot accept. In recent years, the agricultural products and food inspections have been one of the most important issues all over the world. However, traditional food classification based on external features relies on manual processing, which is time consuming and subjective. Invasive detection depends on chemical analysis, and it is expensive, destructive, and experimental samples can never be eaten and used. Hyperspectral imaging is a popular remote sensing technology and widely used in various fields. It uses different materials with different reflection properties to detect different target. In this thesis, we introduce an emerging method to detect Fusarium wilt at the base of Phalaenopsis stems. The detection model divides Phalaenopsis samples into two categories, healthy and infected. The band selection (BS) processing technique based on band prioritization (BP) is applied to extract significant bands and eliminate redundant bands. The algorithms applied are: constrain energy minimization (CEM), spectral information divergence (SID) and SeQuential N-FINDR. These techniques could help detect Fusarium wilt, and we hope the research would help minimize the farmer's losses.
URI: http://hdl.handle.net/11455/98134
文章公開時間: 2018-08-21
Appears in Collections:通訊工程研究所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.