在臺灣，K他命為常見毒品，而其中一種常見的K他命吸食方式為將K他命粉末捲入香菸內後點燃吸食，此種吸食方式造成K他命的氣味在空氣中飄散，嚴重危害人們的居住品質與身體健康。目前並沒有有效且便利的科學方法或工具去立即辨識K菸或是K他命的氣味，由於K他命加熱後會產生一股類似燒塑膠的濃烈氣味，因此本研究利用仿生電子鼻系統，以氣味來辨識K他命。
本研究提出了K他命與K菸氣體定量收集方法，並使用穩定流速之感測系統，讓環境和流速壓力的影響降至最低。利用金屬半導體氧化物感測器陣列感測先前收集之氣體樣本後，藉由訊號前處理，對感測器電阻變化進行特徵擷取，並使用降低維度演算法，使資料複雜度及資料分布之觀察難度降低。透過線性迴歸及分類器，對K他命、不同混合比例之K菸、香菸、K他命混合干擾氣體等氣體進行辨識結果分析及比較。由辨識結果可看出，本系統對K他命、K菸、香菸，三種氣體進行樣本收集及分析後，對於偵測樣本氣體是否含有K他命最高可達到95.92%辨識率；上述三種氣體另外增加含干擾氣體(甲醛)的數種樣本後，對於偵測樣本氣體是否含有K他命最高可達到96.83%辨識率。

Drug abuse has become a serious problem in Taiwan, while ketamine is the most widespread among drugs. There are many methods to take ketamine; among those methods, it is very common in Taiwan that people take ketamine in the way of ketamine cigarettes. By heating up ketamine cigarettes, it produces an odor that smells similar to burning plastic. There are no effective and convenient methods for immediately detecting and identifying the smell of ketamine. Therefore, this research proposed to develop an electronic nose system to identify the gas of ketamine.
In the thesis, standard operating procedures were established for collecting ketamine gas and ketamine cigarette gas at constant flow to quantitatively collect those gas samples least affected by environmental factors. After injecting the gas samples, the variation signal of the sensor array was noted. By feature extraction and dimensionality reduction algorithms, the complexity of data decreased. The classification results were also analyzed through linear regression and classification algorithms. The classification accuracy was up to 95.92% in identifying ketamine, ketamine cigarette, and cigarette; the classification accuracy was up to 96.83% in identifying all gas samples.

摘要 I
Abstract II
致謝 III
目錄 IV
圖目錄 VI
表目錄 VII
第一章 緒論 1
1.1 研究背景與動機 1
1.2 氣體感測系統基本架構 3
1.3 章節簡介 6
第二章 文獻回顧 7
2.1 氣體檢測相關文獻討論 7
2.2 K他命毒品辨識檢測相關文獻討論 8
第三章 實驗系統架構 12
3.1 氣體收集標準化流程 15
3.1.1 K他命和香菸氣體收集流程 15
3.1.2 K菸氣體收集流程 17
3.1.3 K他命氣體與干擾氣體混合流程 18
3.2 氣體實驗流程架設 20
3.2.1 氣體實驗氣路架設 20
3.2.2 氣體反應訊號擷取流程 22
第四章 電子鼻系統訊號處理流程 24
4.1 感測器訊號前處理與特徵擷取 24
4.2 降維演算法 26
4.2.1 主成分分析 (Principal Components Analysis, PCA) 26
4.2.2 線性判別分析 (Linear Discriminant Analysis, LDA) 27
4.3 分類演算法 28
4.3.1 K-近鄰演算法(K-Nearest Neighbors, KNN) 28
4.3.2 交叉驗證(Cross-Validation) 29
4.4 線性迴歸(Linear Regression) 30
第五章 實驗結果與討論 31
5.1 資料種類與誤差 31
5.1.1 各種類氣體資料統整 31
5.1.2 各種類氣體資料平均圖譜與資料誤差 32
5.2 降維演算法處理後資料分佈及辨識結果 35
5.2.1 K他命、K菸、香菸氣體之資料分佈及辨識結果 35
5.2.2 K他命、K菸、香菸、K他命混合干擾氣體之資料分佈及辨識結果 40
5.3 線性迴歸處理結果 45
第六章 結論與未來發展 47
6.1 結論 47
6.2 未來發展 48
參考文獻 49

[1] Foundation for a Drug-Free World (https://www.drugfreeworld.org/drugfacts/prescription/ketamine.html)
[2] Appendix 3 Category three narcotics of Narcotics Hazard Prevention Act; Ministry of Justice; Taiwan: 2017, 06, 14 (https://law.moj.gov.tw/Eng/LawClass/LawContent.aspx?PCODE=C0000008)
[3] Claire A. de March, SangEun Ryu, Gilles Sicard, Cheil Moon, and Jérôme Golebiowski*, Structure–odour relationships reviewed in the postgenomic era, Flavour and Fragrance Journal. 2015, 9, Vol. 30, 342–361
[4] Gardner, J.W.; Bartlett, P.N.; Dodd, G.H.; Shurmer, H.V. Pattern recognition in the Warwick Electronic Nose. In Proceedings of the 8th International Congress of the European Chemoreception Research Organisation, Coventry, UK, 18–22 July 1988.
[5] Harwood, D., Something in the air [electronic nose]. IEE Review 2001, 47, (1), 10-14.
[6] Zhang, L.; Tian, F.; Nie, H.; Dang, L.; Li, G.; Ye, Q.; Kadri, C., Classification of multiple indoor air contaminants by an electronic nose and a hybrid support vector machine. Sensors and Actuators B: Chemical 2012, 174, (0), 114-125.
[7] Dragonieri, S.; van der Schee, M. P.; Massaro, T.; Schiavulli, N.; Brinkman, P.; Pinca, A.; Carratú, P.; Spanevello, A.; Resta, O.; Musti, M.; Sterk, P. J., An electronic nose distinguishes exhaled breath of patients with Malignant Pleural Mesothelioma from controls. Lung Cancer 2012, 75, (3), 326-331.
[8] Dongmin, G.; Zhang, D.; Naimin, L.; Lei, Z.; Jianhua, Y., A Novel Breath Analysis System Based on Electronic Olfaction. Biomedical Engineering, IEEE Transactions on 2010, 57, (11), 2753-2763.
[9] Wilson, A. D.; Baietto, M., Advances in Electronic-Nose Technologies Developed for Biomedical Applications. Sensors 2011, 11, (1), 1105-1176.
[10] Di Natale, C.; Macagnano, A.; Martinelli, E.; Paolesse, R.; Proietti, E.; D’Amico, A., The evaluation of quality of post-harvest oranges and apples by means of an electronic nose. Sensors and Actuators B: Chemical 2001, 78, (1–3), 26-31.
[11] Concina, I.; Falasconi, M.; Sberveglieri, V., Electronic Noses as Flexible Tools to Assess Food Quality and Safety: Should We Trust Them? Sensors Journal, IEEE 2012, 12, (11), 3232-3237.
[12] Macías, M.; Manso, A.; Orellana, C.; Velasco, H.; Caballero, R.; Chamizo, J., Acetic Acid Detection Threshold in Synthetic Wine Samples of a Portable Electronic Nose. Sensors 2012, 13, (1), 208-220.
[13] Ampuero, S.; Bosset, J. O., The electronic nose applied to dairy products: a review. Sensors and Actuators B: Chemical 2003, 94, (1), 1-12.
[14] Berna, A., Metal Oxide Sensors for Electronic Noses and Their Application to Food Analysis. Sensors 2010, 10, (4), 3882-3910.
[15] Baldwin, E. A.; Bai, J.; Plotto, A.; Dea, S., Electronic Noses and Tongues: Applications for the Food and Pharmaceutical Industries. Sensors 2011, 11, (5), 4744-4766.
[16] Hasan, N.; Ejaz, N.; Ejaz, W.; Kim, H., Meat and Fish Freshness Inspection System Based on Odor Sensing. Sensors 2012, 12, (11), 15542-15557.
[17] Gardner, J. W.; Shin, H. W.; Hines, E. L.; Dow, C. S., An electronic nose system for monitoring the quality of potable water. Sensors and Actuators B: Chemical 2000, 69, (3), 336-341.
[18] Baby, R. E.; Cabezas, M.; Walsöe de Reca, E. N., Electronic nose: a useful tool for monitoring environmental contamination. Sensors and Actuators B: Chemical 2000, 69, (3), 214-218.
[19] J. Goschnick;M. Harms Landmine, Detection with an Electronic Nose Mounted on an Airship. NATO Science Series 2002, 66, (0), pp 83-91.
[20] Wilson, A. D., Review of Electronic-nose Technologies and Algorithms to Detect Hazardous Chemicals in the Environment. Procedia Technology 2012, 1, (0), 453-463.
[21] Tsow, F.; Forzani, E.; Rai, A.; Rui, W.; Tsui, R.; Mastroianni, S.; Knobbe, C.; Gandolfi, A. J.; Tao, N. J., A Wearable and Wireless Sensor System for Real-Time Monitoring of Toxic Environmental Volatile Organic Compounds. Sensors Journal, IEEE 2009, 9, (12), 1734-1740.
[22] Moncrieff, R. W., An instrument for measuring and classifying odours, Journal of Applied Physiology 1961, 16, (4), 742-749.
[23] BIRD, Robert C. Examination of the Training and Reliability of the Narcotics Detection Dog, An. Ky. LJ, 1996, 85(405).
[24] MYERS, Richard. Detector Dogs and Probable Cause. George Mason Law Review, 2006, 14.1.
[25] Frank C. Zee, Micro-Electronic Nose System, 2011, Publication Number: AAT 3472649; ISBN: 9781124870069
[26] Science Study Monthly, 2017, 4, No.56-04, 13-22
[27] Regulations Governing Drug Abuse Urine Testing Operations, Ministry of Health and Welfare, Taiwan, 2014, 10, 23 (https://law.moj.gov.tw/ENG/LawClass/LawAll.aspx?pcode=L0030044)
[28] Debra A. K., Corinne M., Gunnar D. N., Peder W., Chapter 3. Formaldehyde, WHO Guidelines for Indoor Air Quality: Selected Pollutants, 2010.
[29] Figaro, Gas Sensors & Modules (https://www.figaro.co.jp/en/product/sensor/)
[30] TGS-2600 datasheet
(https://www.figaro.co.jp/en/product/docs/tgs2600_product_information_rev02.pdf)
[31] Karl Pearson. "On lines and planes of closest fit to systems of points in space". Philosophical Magazine Series 1901, 6, 2(11), 559–572,
[32] Fisher, R. A. "The Use of Multiple Measurements in Taxonomic Problems". Annals of Eugenics, 1936, 7(2), 179–188
[33] Altman, N. S. An introduction to kernel and nearest-neighbor nonparametric regression, The American Statistician. 1992, 46 (3), 175–185.
[34] Kohavi, Ron. A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, 1995, 2 (12)