稀疏表示法近年來在影像分析上有很好的辨識效果，且基於稀疏表示的語者識別方法相繼被提出，本論文藉由稀疏表示線性組合的特性進行氣味的分析，目標在於做單一氣味以及混合氣味的辨識，首先，選用生活中常見的20種氣味源(混合物)，建立稀疏表示演算法進行單一氣味的辨識，接著，將兩種單一氣味組合成18種混合氣味，訓練資料庫以單一氣味的數據資料作為基礎，使用稀疏表示法的特性進行混合氣味的辨識，期望可以將混合氣味判斷為單一氣味的線性組合。
在辨識氣味成分前，本論文提出以稀疏表示的特性判斷氣味源為單一氣味或是混合氣味，正確率可達到7成以上。接著，為了獲取有意義的多變量響應作為特徵進行氣味識別，我們分別在資料前處理、特徵選擇、分類器以及辨識方法上進行比較與分析，找到最佳的資料處理以及分類方法。資料前處理方法上，使用不同的壓縮方式以及正規化；特徵選擇包含特徵選取以及特徵萃取，前者採用循序向後的特徵選擇演算法，後者則使用了主成分分析方法以及線性識別分析；分類器選用最近鄰居分類器以及稀疏表示法；辨識方法上，在單一氣味的分析上使用多類別辨識方法，混合氣味則比較多類別辨識方法以及多標籤辨識方法的分析結果。本論文中的混合氣味資料共有18組，每一組是由兩種單一氣味所組成，因此共有36個目標氣味需要判別，目前最多可以判斷出14個目標氣味，其中包含正確地判斷兩組混合氣味。

Sparse Representation Classification (SRC) has performed well in the field of image analysis and speaker identification. In this thesis, we applied SRC in single and mixed odor recognition. First, we chose 20 kinds of odor sources and built an SRC-based algorithm to recognize them. Then, we produced 18 kinds of mixed odors by mixing two of the 20 kinds of odor sources. We attempted to recognize the mixed odors by learning from the training data set which consists of single odor data. A mixed odor could be recognized as a linear combination of the single odors by using SRC.
Before performing odor recognition, we detected whether the sample was a single odor or a mixed odor with >70% accuracy using an SR-based method. Then, odors were analyzed by the following steps: data preprocessing, dimension reduction, classification and recognition, respectively. For data preprocessing, various methods were applied to compress and normalize the raw data. Afterwards, dimension reduction was achieved via feature selection and feature extraction. We used Sequential Backward Selection (SBS) for feature selection, and Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) for feature extraction. For data classification, we applied K-Nearest Neighbor Classification (KNNC) and SRC. For recognition, we used multiclass classification for single odors identification, and compared the results of mixed odor identification produced by a multiclass algorithm and a multi-label algorithm. The mixed-odor dataset consisted of 18 pairs of odors, so there were 36 targets to identify. Results show that, as many as 14 targets could be successfully identified.

[1] D. Purves, G. J. Augustine, and D. Fitzpatrick, Neuroscience, 5th Edition. Sinauer Associates, Inc., 2012.
[2] J. W. Gardner and P. N. Bartlett, “A brief history of electronic noses,” Sensors Actuators B Chem., vol. 19, pp. 211–220, 1994.
[3] D. W. Ballantine, R. M. White, and S. J. Martin, Acoustic Wave Sensors: Theory Design and Physico-Chemical Applications. Academic Press, 1997.
[4] J. W. Gardner, Handbook of Machine Olfaction. WILEY-VCH, 2003.
[5] P. C. Jain and R. Kushwaha, “Wireless gas sensor network for detection and monitoring of harmful gases in utility areas and industries,” 2012 Sixth Int. Conf. Sens. Technol., pp. 642–646, 2012.
[6] K.-T. Tang, S.-W. Chiu, C.-H. Shih, C.-L. Chang, C.-M. Yang, D.-J. Yao, J.-H. Wang, C.-M. Huang, H. Chen, K.-H. Chang, C.-C. Hsieh, T.-H. Chang, M.-F. Chang, C.-M. Wang, Y.-W. Liu, T.-J. Chen, C.-H. Yang, H. Chiueh, J.-M. Shyu, “A 0.5V 1.27mW Nose-on-a-Chip for Rapid Diagnosis of Ventilatorassociated Pneumonia,” the 2014 IEEE international Solid-State Circuits Conference (ISSCC 2014), pp. 420–421, 2014
[7] 許柏安, 「快速混合氣體辨識方法之研究」, 國立清華大學資訊工程研究所, 碩士論文, 2012.
[8] J. Yang, J. Wright, T. S. Huang, and Y. Ma, “Image super-resolution via sparse representation,” TIP, 19(11):2861-2873, 2010.
[9] L Tan, A. Alwan, G Kossan, ML. Cody, and CE. Taylor, “Dynamic time warping and sparse representation classification for birdsong phrase classi- fication using limited training data”. J Acoust Soc Am, 137(3):1069–1080, 2015.
[10] I. Naseem, R. Togneri, and M. Bennamoun, “Sparse Representation for Speaker Identification,” in Proc. of ICPR, pp. 4460-4463, 2010.
[11] J. Wright , A. Yang , A. Ganesh , S. Sastry and Y. Ma “Robust face recognition via sparse representation”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 30, no. 2, pp.210 -227, 2009
[12] 王光耀, 「基於稀疏表示之語者辨識之研究」, 國立中央大學資訊工程學系, 碩士論文, 2013.
[13] “Figaro Sensor.” Retrieved: http://www.figarosensor.com/.
[14] R. Gutierrez-Osuna and H. T. Nagle, “A method for evaluating data-preprocessing techniques for odor classification with an array of gas sensors,” IEEE Trans. Syst. Man, Cybern. Part B Cybern., vol. 29, no. 5, pp. 626–632, 1999.
[15] NIST/SEMATECH e-Handbook of Statistical Methods, http://www.itl.nist.gov/div898/handbook/, 2003.
[16] M. K. Muezzinoglu, A. Vergara, R. Huerta, N. Rulkov, M. I. Rabinovich, A. Selverston, and H. D. I. Abarbanel, “Acceleration of chemo-sensory information processing using transient features,” Sensors Actuators B Chem., vol. 137, no. 2, pp. 507–512, Apr. 2009.
[17] S. Marco and A. Gutiérrez-gálvez, “Signal and Data Processing for Machine Olfaction and Chemical Sensing : A Review,” vol. 12, no. 11, pp. 3189–3214, 2012.
[18] M. Dash and H. Liu, “Feature selection for classification,” Intell. Data Anal., vol. 1, no. 3, pp. 131–156, 1997.
[19] 王家銘, 「利用樣式識別實現電子鼻肺炎偵測」, 國立清華大學電機工程研究所, 碩士論文, 2013.
[20] G. John, R. Kohavi, and K. Pfleger, “Irrelevant features and the subset selection problem.” ICML, pp. 121–129, 1994.
[21] C. C. Reyes-Aldasoro and a. Bhalerao, “The Bhattacharyya space for feature selection and its application to texture segmentation,” Pattern Recognit., vol. 39, no. 5, pp. 812–826, 2006.
[22] A. M. Martinez and A. C. Kak, “PCA versus LDA,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 23, no. 2, pp. 228–233, 2001.
[23] A. Fort, N. Machetti, S. Rocchi, M. B. Serrano Santos, L. Tondi, N. Ulivieri, V. Vignoli, and G. Sberveglieri, “Tin oxide gas sensing: Comparison among different measurement techniques for gas mixture classification,” IEEE Trans. Instrum. Meas., vol. 52, no. 3, pp. 921–926, Jun. 2003.
[24] I.T. Jolliffe, Principal Component Analysis, 2nd ed, vol. 98. Springer-Verlag, 2002.
[25] 周志成(2013)。主成分分析。民104年8月29日，取自：https://ccjou.wordpress.com/2013/04/15/%E4%B8%BB%E6%88%90%E5%88%86%E5%88%86%E6%9E%90/
[26] H. Y. Jie, H. Yu, and J. Yang, “A direct LDA algorithm for high-dimensional data-with application to face recognition,” Pattern Recognit., vol. 34, no. 10, pp. 2067–2070, 2001.
[27] 周志成(2014)。費雪的判別分析與線性判別分析。民104年8月29日，取自：https://ccjou.wordpress.com/2014/03/14/%E8%B2%BB%E9%9B%AA%E7%9A%84%E5%88%A4%E5%88%A5%E5%88%86%E6%9E%90%E8%88%87%E7%B7%9A%E6%80%A7%E5%88%A4%E5%88%A5%E5%88%86%E6%9E%90/
[28] Y. Shin, S. Lee, J. Lee, and H.-N. Lee, “Sparse representation-based classification scheme for motor imagery-based brain–computer interface systems,” Journal of Neural Engineering, vol. 9, no. 5. p. 056002, 2012.
[29] D. L. Donoho, “Compressed sensing,” IEEE Trans. Inf. Theory, vol. 52, no. 4, pp. 1289–1306, 2006.
[30] R. G. Baraniuk, “Compressive sensing,” IEEE Signal Process. Mag., vol. 24, no. 4, p. 118, 2007.
[31] E. J. Candes and M. B. Wakin, “An introduction to compressive sampling,” IEEE Signal Process. Mag., vol. 25, no. 2, 2008.
[32] E. J. Candès, J. K. Romberg, and T. Tao, “Stable signal recovery from incomplete and inaccurate measurements,” Commun. Pure Appl. Math., vol. 59, no. 8, pp. 1207–1223, 2006.
[33] M. Elad, “Sparse and Redundant Representations,” Sparse Redundant Represent. From Theory to Appl. Signal Image Process., pp. 359–361, 2010.
[34] S. Gao, I. W. H. Tsang, and L. T. Chia, “Kernel sparse representation for image classification and face recognition,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6314 LNCS, no. PART 4, pp. 1–14, 2010.
[35] M. Zhang and Z. Zhou, “A k-Nearest Neighbor Based Algorithm for Multi-label Classification,” 2005 IEEE Int. Conf. Granul. Comput., vol. 2, pp. 718–721 Vol. 2, 2005.
[36] R. Kohavi, “A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection,” in International Joint Conference on Artificial Intelligence, 1995, vol. 14, no. 12, pp. 1137–1143.
[37] R. Gutierrez-Osuna, “Pattern analysis for machine olfaction: A review,” IEEE Sensors Journal, vol. 2, no. 3. pp. 189–202, 2002.
[38] M. Aly, “Survey on multiclass classification methods extensible algorithms,” Neural Networks, no. November, pp. 1–9, 2005.
[39] L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stone, Classification and Regression Trees, vol. 19. Belmont, California: Wadsworth, 1984.
[40] J. R. Quinlan, C4.5: Programs for Machine Learning, vol. 1, no. 3. Morgan Kaufmann Publishers Inc. San Francisco, CA, USA, 1993.
[41] C. M. Bishop, Neural Networks for Pattern Recognition, vol. 92. Oxford University Press, USA, 1995.
[42] S. D. Bay, “Combining nearest neighbor classifiers through multiple feature subsets,” in Proceedings of the 17th International Conference on Machine Learning, pp. 37–45, 1998.
[43] I. Rish, “An empirical study of the naive Bayes classifier,” IJCAI 2001 Work. Empir. methods Artif. Intell., vol. 22230, pp. 41–46, 2001.
[44] C. Cortes and V. Vapnik, “Support-vector networks,” Mach. Learn., vol. 20, no. 3, pp. 273–297, 1995.
[45] C. J. C. Burges, “A tutorial on support vector machines for pattern recognition,” Data Min. Knowl. Discov., vol. 2, pp. 121–167, 1998.
[46] E. A. Cherman, M. C. Monard, J. Metz , “Multi-label problem transformation methods: a case study,” CLEI Electron. J., vol. 14, no. 4, 2011.
[47] C. Z. S. Z. Tao L., “Empirical studies on multi-label classification,” in Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI, 2006, pp. 86–89.
[48] K. Trohidis and G. Kalliris, “Multi-label classification of music into emotions,” Learning, vol. 2008, pp. 325–330, 2008.
[49] M. R. Boutell, J. Luo, X. Shen, and C. M. Brown, “Learning multi-label scene classification,” Pattern Recognit., vol. 37, no. 9, pp. 1757–1771, 2004.
[50] M. Zhang and Z. Zhou, “Multi-label learning by instance differentiation,” AAAI, pp. 669–674, 2007.
[51] 楊廷然, 「利用多標籤分類器實現電子鼻混合氣體識別方法之研究」, 國立清華大學電機工程研究所, 碩士論文, 2014.
[52] G. Tsoumakas, I. Katakis, and I. Vlahavas, “Mining multi-label data,” Data Min. Knowl. Discov. Handb., pp. 667–685, 2010.
[53] J. Read, B. Pfahringer, G. Holmes, and E. Frank, “Classifier chains for multi-label classification,” Mach. Learn., vol. 85, no. 3, pp. 333–359, Jun. 2011.