台灣地處緯度與地形落差大，造成農業發展的多樣性，舉凡食用的農產品以及觀賞用花卉植物發展出傲人之農業成就。但因環境多樣造成病害亦多，諸如蝴蝶蘭，其病毒困擾造成栽培者的巨大損失，尤其病毒有潛伏期，以現行的檢測方法，除了以栽種人員憑經驗以外觀判斷花株是否感染以外，就是耗時費力的生物檢測，例如：酵素免疫吸附分析法、聚合酶連鎖反應、免疫試紙、磁減量法等。篩選檢測無法普及，僅能抽樣，故易有漏網之魚。
本技術以手持光學探測方式用LED發光照射葉子再以探測頭收光，汲取訊號後經光譜儀還有控制板整理光波數據，搭配手機回傳至資料庫。將資料經平滑化、標準化及PCA降維前處理後，使用『支持向量機』及『極限梯度提升』兩種演算法進行分類分析，得出對於蘭花病毒的判斷準確度及效果。其中SVC判斷生病蘭花的準確率達87%、XGBoost判斷生病蘭花的準確率達91%(不論是已經發病還是未發病處於潛伏期的苗株都一體適用)，相對於以往耗時費力且無法普及到每一株蘭花的方式，本遠距、無耗材、手持式的蘭花病毒快篩機，讓花農能在其肉眼看不到的潛伏期得知感染結果，以防範群聚感染。

Because a large gap showed in the location latitude and topography, the diversity of agricultural development, including edible agricultural products and ornamental flowers and plants, in Taiwan have developed proud agricultural achievements. However, diverse environments also resulted in many diseases. For example, the virus troubles of phalaenopsis cause huge losses to the growers. Because the virus has an incubation period, the discrimination of the infection for orchid plants is regularly by the visual experience of the planter, or the time-consuming and laborious biological tests, such as enzyme link immunosorbent assay (ELISA), polymerase chain reaction (PCR), immuno-strip detection kit, magnetic Immunomagnetic reduction (IMR), etc. Based on these technologies, the screening cannot be popularized as the full inspection, and only be as sampling inspection with the missing risks.
This technology based on a hand-held optical detection method to illuminate the leaves with LED light, and then receives the fluorescent light with the probe. After receiving the signal, the light wave data is sorted by the spectrometer and the control board, and it is returned to the database with the mobile phone. After all the data profile is smoothed, standardized and pre-processed by PCA for dimensionality reduction, two algorithms, i.e. support Vector Machine (SVM) and extreme gradient boosting (eXtreme Gradient Boosting, XGBoost), were used for classification and analysis. The accuracy and effect of the judgment. Among them, SVC has an accuracy rate of 87% in judging diseased orchids, and XGBoost has an accuracy rate of 91% in judging diseased orchids (whatever visible symptoms or not). In comparison with the time-consuming and labor-intensive mentioned methods, this remote, non-consumable, handheld, rapid screening machine allowed the flower grower to detect the infection results early than the visible symptoms to prevent the cluster infection.

[1] 農業生技資訊加值小組,”全球蝴蝶蘭市場發展與趨勢”,台灣經濟研究院生物科技產業研究中心,1-2.7.17,2011。
[2] 謝廷芳、張義璋、杜金池. 2 0 0 0. 由Bipolaris zizaniae 引起的文心蘭花瓣斑點病. 植病會刊9 : 7 5 - 7 8.
[3] 安寶貞, 位國慶, 張清安, 童伯開, 黃德昌, 楊秀珠, 2001. 病毒病害。p.12-56. 植物保護圖鑑系列-洋蘭保護。行政院農業委員會動植物防疫檢疫局出版。
[4] 張清安 1994. 蘭花病毒病之特性與防治。p.129-148. 蘭花經濟栽培技術。賴本智等主編。行政院青年輔導委員會創業輔導叢書三-17。行政院青年輔導委員會出版。148pp。
[5] 曾哲明, 免疫學Immunology, 初版, 新文京開發出版,台灣新北市,p.110-112 ,2005
[6] 詹富智,盧耀村,陳慶忠,” 植物檢疫病毒偵測技術”, 植物重要防疫檢疫病害診斷鑑定技術研習會專刊,(二),p.43-44,2011
[7] 吳凱智,免疫層析試紙測試法組裝最適化對streptavidin蛋白質之快速檢測,中原戴學化學工程學系碩士學位論文,2007
[8] S.Y. Yang, Z.F. Jian, J.J. Chieh, H.E. Horng, H.C. Yang, I.J. Huangc, Chin-Yih Hong, “Wash-free, antibody-assisted magnetoreduction assays of orchid viruses”, ScienceDirect, Journal of Virological Methods149,p.334-337,2008
[9] 陳健尉,生物晶片：技術發展及在生命科學上之應用, 「進階生物技術」第十五章(教育部生物技術教育改進計劃),2004
[10] Doreen Thoren, Peter Thoren, Urs Schmidhalter,(2010). Influence of ambient light and temperature on laser-induced hlorophyll fluorescence measurements. Europ. J. Agronomy 32 (2010) 169–176.
[11]Hopkins, W.G. and P.A. Hüner. 徐善德、廖玉琬編譯. 2006. 植物生理學. 偉明圖書有限公司. 台北. 台灣.
[12]潘瑞熾. 2008. 植物生理學. 藝軒圖書出版社. 台北. 台灣.
[13]Texas A&M University. http://generalhorticulture.tamu.edu/lectsupl/Physiol/physiol.html.
[14]CIBA Foundation Symposium, Chlorophyll Organization and Energy Transfer in Photosynthesis. Novartis Foundation Symposia 第915卷, John Wiley & Sons(2009).
[15] C. Gameiroa, A.B. Utkin, P. Cartaxana, J. Marques da Silva, A.R. Matos(2015). The use of laser induced chlorophyll fluorescence (LIF) as a fast and
non-destructive method to investigate water deficit in Arabidopsis. Agricultural Water Management 164(2016)127-136.

[16]William H. Press, and Saul A. Teukolsky(1990). Savitzky-Golay Smoothing Filters. Computers in Physics 4, 669(1990); doi: 10.1063/1.4822961.

[17] Wu,Tom(2018,Decemeber 18). 資料的正規化(Normalization)及標準化(Standardization). Retrieved July 29, 2020, from https://aifreeblog.herokuapp.com/posts/54/data_science_203/

[18] Chih-Sheng Huang (2018, April 20). 機器/統計學習:主成分分析(Principal Component Analysis, PCA). Retrieved July 27, 2020, from

[19]N.J. Nalini, S. Palanivel(2015). Music emotion recognition: The combined evidence of MFCC and residual phase. Egyptian Informatics Journal 17(2016)1-10.

[20] Tianqi Chen, Carlos Guestrin(2016). XGBoost: A Scalable Tree Boosting System. KDD '16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)785-794.