本研究利用影像處理技術自動化監測不同種類之果蠅群組互動行為，藉由改良子區塊分割的空間金字塔配對與配合投票法，以進行四種果蠅種類辨識，並達到100%的辨識正確率；此外果蠅追蹤能夠實現10隻果蠅同時監測且獨立分析取得果蠅之位置、大小及頭部方向，監測時的果蠅配對錯誤率更是小於0.01%。因此本研究發展之不同種類之果蠅群組互動監測技術能夠有效地降低實驗的人力成本與節省大量時間。
此外，而本研究也提出Drosophila melanogaster於不同種類之果蠅群組互動行為與同種間的互動行為之比較。藉此研究法，發現Drosophila melanogaster於其他三種類的果蠅群組內之平均互動次數較其他種類高，可是並無明顯差異。

This research uses image processing techniques to automatically monitor interactions within groups of different categories of Drosophila. With revised sub-regions of Spatial Pyramid Matching and winner take all method, we achieve 100% recognition rate in four categories of Drosophila. In addition, through drosophila tracking, we are able to monitor 10 Drosophila simultaneously and analyze every individual drosophila to acquire its position, size and heading angle with a rate of mismatching sequential drosophila under 0.01%. Therefore, the interactions within groups of interspecies monitoring which this research developed would efficiently enhance the experiment performance by reducing labor cost and time consumption.
This research proposes the comparison between interspecies and within-species interactions within groups. With the outcome of the interspecies interaction experiment, we come to a conclusion that the average interactions of drosophila melanogaster is higher than the other species of drosophila but there is no statistical significance.

摘要 I
Abstract II
致謝 III
目錄 V
圖目錄 IX
表目錄 XIII
第一章 緒論 1
1.1 前言 1
1.2 研究動機與論文架構 2
第二章 文獻回顧 4
2.1 影像追蹤 4
2.1.1 背景消去法(Background Subtraction) 4
2.1.2 瞬間差分法 5
2.2 果蠅行為分析與行為追蹤 5
2.3 影像分類 10
2.3.1 興趣點偵測與描述 10
A. 興趣點偵測 10
B. 興趣點描述 11
2.3.2 向量量化編碼(Vector Quantization) 12
2.3.3 特徵表示方法 12
A. 特徵詞袋模型(Bag-of-features) 12
B. 空間金字塔對照(Spatial Pyramid Matching，SPM) 14
2.4 分類器 16
第三章 研究方法 18
3.1 實驗裝置 20
3.2 果蠅位置追蹤 23
3.2.1 K-means演算法 26
3.2.2 果蠅位置預測與配對模型 29
3.3 果蠅種類辨識 32
3.3.1 興趣點偵測與描述 32
3.3.2 建立字庫 40
3.3.3 特徵表示方法 40
3.3.4 分類器 44
3.4 果蠅互動行為記錄技術 50
第四章 結果與討論 55
4.1 果蠅位置追蹤 57
4.2 果蠅種類辨識 60
4.3 果蠅互動行為分析 69
4.3.1 同種類的互動分析 69
4.3.2 D.M.於不同種類之果蠅互動行為分析 71
第五章 結論 75
5.1 本研究之貢獻 75
5.2 未來展望 76
參考文獻 78
附件一、5隻同種果蠅互動表 82
附件二、5隻同種果蠅於平台內之移動軌跡圖與四象限分佈機率圖 84
附件三、10隻同種果蠅互動表 86
附件四、10隻同種果蠅於平台內之移動軌跡圖與四象限分佈機率圖 90
附件五、不同種果蠅之互動數據表 94

[1] S. Y. Elhabian, K. M. El-Sayed* and S. H. Ahmed, “Moving object detection in spatial domain using background removal techniques - State-of-Art,” Recent Patents on Computer Science, vol.1, pp. 32-54, 2008.
[2] JR. Martin “A portrait of locomotor behaviour in Drosophila determined by a video-tracking paradigm,” Behavioural Processes, vol. 67, pp. 207-219, 2004.
[3] N. Dimitrijevic, S. Dzitoyeva and H. ManevAn, “Automated assay of the behavioral effects of cocaine injections in adult Drosophila,” Journal of Neuroscience Methods, vol. 137, pp. 181-184, 2004.
[4] R. B. Ramazani, H. R. Krishnan, S. E. Bergeson and N. S. Atkinson, “Computer automated movement detection for the analysis of behavior,” Journal of Neuroscience Methods, vol. 162, pp. 171-179, 2007.
[5] K. Branson, A. Robie, J. Bender, P. Perona and M. H. Dickinson, “High-throughput ethomics in large groups of Drosophila,” Nature Methods, vol. 6, pp. 451-457, 2009.
[6] J. Schneidera, M. H. Dickinsonb and J. D. Levinea, “Social structures depend on innate determinants and chemosensory processing in Drosophila,” Proceedings of the National Academy of Sciences, vol. 109, pp. 17174-17179, 2012.
[7] P. Fan, D. S. Manoli, O. M. Ahmed, Y. Chen, N. Agarwa, S. Kwong, A. G. Cai, J. Neitz, A. Renslo, B. S. Baker, and N. M. Shah, “Genetic and neural mechanisms that inhibit Drosophila from mating with other species,” Cell, vol. 154, pp. 89-102, 2013.
[8] K. Mikolajczyk, B. Leibe, and B. SchieleLocal, “Features for Object Class Recognition,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1792–1799, 2005.
[9] K. Mikolajczyk and C. Schmid, “Scale & affine invariant interest point detectors,” International Journal of Computer Vision (IJCV), vol. 60, pp. 63–86, 2004.
[10] C. Harris and M. Stephens, “A Combined Corner and Edge Detector”, Proceedings of 4th Alvey Vision Conference, pp. 147-151, 1988.
[11] D. Lowe, “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision (IJCV), vol. 60, pp. 91–110, 2004.
[12] T. Kadir and M. Brady, “Scale, Saliency and Image Description,” International Journal of Computer Vision (IJCV), vol. 45, pp. 83–105, 2001.
[13] J. Matas, O. Chum, M. Urban, and T. Pajdla. “Robust wide baseline stereo from maximally stable extremal regions,” Proceedings of the British Machine Vision Conference (BMVC), pp. 36.1–36.10, 2002.
[14] K. Mikolajczyk and C. Schmid, “A performance evaluation of local descriptors,” IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 27, pp. 1615-1630, 2005.
[15] J. R. R. Uijlings, A. W. M. Smeulders, and R. J. H. Scha, “Real-time visual concept classification,” IEEE Transactions on Multimedia, vol. 12, pp. 665-681, 2010.
[16] J. B. MacQueen, "Some Methods for classification and Analysis of Multivariate Observations," Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, vol. 1, pp. 281-297, 1967.
[17] K. E. A. Van De Sande, T. Gevers, and C. G. M. Snoek, “Empowering visual categorization with the GPU,” IEEE Transactions on Multimedia, vol. 13, pp. 60-70, 2011.
[18] L. Breiman, "Random Forests," Machine Learning, vol. 45, pp. 5- 32, 2001.
[19] J. Philbin, O. Chum, M. Isard, J. Sivic, and A. Zisserman, “Object retrieval with large vocabularies and fast spatial matching,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1-8, 2007
[20] D. Nist´er and H. Stew´enius, “Scalable Recognition with a Vocabulary Tree,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2161-2168, 2006.
[21] G. Csurka, C. R. Dance, L. Fan, J. Willamowski and C. Bray, ” Visual Categorization with Bags of Keypoints,” Workshop on statistical learning in computer vision, pp. 1-22, 2004.
[22] J. Sivic and A. Zisserman, “Video google: a text retrieval approach to object matching in videos,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1470-1477, 2003.
[23] L. Zhu, A. Rao and A. Zhang, “Theory of keyblock-based image retrieval,” ACM Transactions on Information Systems, vol. 20, pp. 224-257, 2002.
[24] H. Jegou, M. Douze and C. Schmid, “Packing bag-of-features,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 2357-2364, 2009.
[25] S. Lazebnik, C. Schmid and J. Ponce, “Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2169-2178, 2006.
[26] K. Grauman and T. Darrell, “The Pyramid Match Kernel: Discriminative Classification with Sets of Image Features,” Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1458-1465, 2005.
[27] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273-297, 1995.
[28] T. Cover and P. Hart, “Nearest Neighbor Pattern Classification,” IEEE Transactions on Information Theory, pp. 21-27, 1967.
[29] J. R. Quinlan, “Induction of Decision Trees,” Machine Learning, pp. 81-106, 1986.
[30] Y. Yang and X. Liu, “A Re-Examination of Text Categorization Methods,” ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 42-49, 1999.
[31] J. Kim, B. Kim and S. Savarese, “Comparing Image Classification Methods: K-Nearest-Neighbor and Support-Vector-Machines,” Proceedings of the 2012 American conference on Applied Mathematics, pp. 133-138, 2012.
[32] C. Demirkesen and H. Cherifi, “A Comparison of Multiclass SVM Methods for Real World Natural Scenes,” International Conference on Advanced Concepts for Intelligent Vision Systems, 2008.
[33] R. Zhao, W. Ouyang and X. Wang, "Unsupervised Salience Learning for Person Re-identification," IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3586-3593, 2013.