跳到主要內容

臺灣博碩士論文加值系統

(100.28.231.85) 您好!臺灣時間:2024/11/14 09:41
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:王子睿
研究生(外文):Tzu-Jui Wang
論文名稱:應用於地點影像辨識之快速局部L-SVMs群體分類器
論文名稱(外文):Efficient Ensemble of Local L-SVMs for Visual Place Categorization
指導教授:唐之瑋
指導教授(外文):Chih-Wei Tang
學位類別:碩士
校院名稱:國立中央大學
系所名稱:通訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2011
畢業學年度:99
語文別:中文
論文頁數:67
中文關鍵詞:決策融合支撐向量機地點辨識
外文關鍵詞:decision fusionSVMPlace categorization
相關次數:
  • 被引用被引用:0
  • 點閱點閱:239
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
近年來基於影像之地點辨識(Visual place categorization)是研究人員重視的研究議題。然而,一影像辨識系統常受到物件大小、光線條件、物件遮蔽、視角變化影響。此外,地點辨識系統必須具備有效率的訓練及測試過程。相關的地點辨識研究中,其在分類影像內特徵時,大多採用非線性支撐向量機(support vector machine, SVM),因其相較線性 SVM普遍能獲得較佳的辨識率,但其缺點是訓練階段之時間複雜度為O(N2)~O(N3),N為訓練樣本數量,如此無法有效率地處理大規模(large-scale)樣本,因此本論文提出基於局部特性之linear SVM之群體分類器(ensemble of local linear SVMs, ELL-SVMs),可有效降低訓練階段之時間複雜度至O(N1.5)。
  本系統分為訓練階段及測試階段。於訓練階段,我們提出一創造群體(ensemble creation)的方法,其可在原特徵空間中找尋近乎線性可分割(linearly-separable)之子集,再利用一linear SVM分割每個子集內的樣本。在測試階段,對於影像中各個所偵測的特徵,我們利用貝式分類器及最近鄰居法則(nearest neighbor rule)個別選擇對每個特徵最佳的分類器。接著,本論文提出基於信心度之加權式一對多方案(confidence-based weighted one-against-all, CW-OAA)以結合影像中各個特徵之分類結果,以決定影像所屬地點。實驗結果顯示,本論文所提出的ELL-SVMs在訓練速度上,相似於FaLK-SVM,且優於BVM,在測試速度上,同時優於FaLK-SVM、BVM以及標準SVM。此外,本論文所提出之整合式地點影像的辨識率,皆優於FaLK-SVM、BVM以及標準SVM搭配OAA之方案。
Recently visual place categorization is an important research topic due to its numerous potential applications. However, such visual categorization system is easily affected by object scale, illumination conditions, object occlusion and viewpoints. In addition, categorization system should be efficiently trained and tested with huge amount of visual cues extracted in a very short period. Relevant researches on visual place categorization rely on non-linear SVM to categorize those visual cues within each image, since non-linear SVM has always shown promising categorization results. However, its major defect is that it suffers from O(N2) to O(N3) in training complexity and O(D‧S) in test complexity, where N is the size of training data, D is the dimension of data vector, and S is the amount of support vector. Efficient training and test pro-cesses are demanding for tackling large-scale categorization problems. Therefore, this thesis proposes ensemble of local linear SVMs (ELL-SVMs), lowering training com-plexity to O(N1.5).
Our proposed scheme has training and test phases. In training phase, we propose a scheme for generating ensemble of local linear SVM (L-SVM). This idea is derived from discovering the linearly-separable partitions among training data, while such partitions are found, those partitions could be classified by linear SVM instead of non-linear one. In test phase, we impose nearest neighbour rule into Bayes decision rule to assist in identifying the best trained local L-SVM for test sample. Afterwards, we further propose confidence-based weighted one-against-all (CW-OAA) approach to fuse the categorization results of visual cues within an image, and thus to categorize the image. Empirically, the training speed of proposed ELL-SVMs is similar to that of FaLK-SVM and much lower than those of BVM and standard SVM. The test speed of ELL-SVMs is lower than FaLK-SVM, BVM and standard SVM. Moreover, the categorization ability of ELL-SVMs with CW-OAA outperforms three SVM’s variants with OAA.
摘要 I
Abstract III
致謝 VI
第一章 緒論 1
1.1 前言 1
1.2 研究動機 1
1.3 研究方法 3
1.4 論文架構 4
第二章 地點辨識演算法之文獻回顧 5
2.1 典型辨識系統之架構 5
2.2 應用於地點辨識之影像特徵擷取 7
2.3 以視覺為基礎之地點辨識演算法現況 11
2.4 總結 13
第三章 Support Vector Machine (SVM)相關技術介紹 14
3.1 Support Vector Machine (SVM) 14
3.2 多類別Support Vector Machine (Multi-Class SVM) 22
3.3 Linear Support Vector Machine之快速訓練方案現況 23
3.4 Kernelized Support Vector Machine之快速訓練方案現 23
3.5 總結 25
第四章 本論文所提出之地點影像分類演算法 27
4.1 系統流程概述 27
4.2 訓練階段 28
4.2.1 貪婪式分割訓練樣本(Greedily Partitioning Training Data) 28
4.2.2 量化已分割樣本(Quantizing Partition using K-means++) 31
4.2.3 產生群體局部L-SVMs (Generating Ensemble of L-SVMs) 32
4.3 測試階段 35
4.3.1 基於貝式決策法(Bayesian Decision Rule)與最近鄰居(Nearest Neighbor Rule)之分類器選擇 35
4.3.2 近似最近鄰居距離(Approximated Nearest Neighbor 38
4.3.3 加權式決策值 (Weighted Decision Value) 39
4.3.4 基於信心度之加權式一對多決策法(Confidence-based Weighted One-against-All Scheme) 45
4.4 總結 47
第五章 實驗結果與討論 48
5.1 實驗環境及測試影像資料庫 48
5.2 分類SIFT描述子之效能分析 50
5.2.1 訓練及測試時間分析 52
5.2.2 辨識率分析 55
5.3 地點影像辨識方案分析 60
5.4 總結 64
第六章 結論及未來展望 66
參考文獻 67
[1]H. D. Whyte and T. Bailey, “Simultaneous localization and mapping: Part I,” IEEE Robotics & Automation Magazine, Vol. 13, No. 2, pp.99-110, Jun. 2006.
[2]B. E. Boser, I. Guyon, and V. Vapnik, “A training algorithm for optimal margin classifiers,” in Proceedings of ACM Conference on Learning Theory, Jul. 1992.
[3]I. W. Tsang, J. T. Kwok, and P.-M. Cheung, “Core vector machines: Fast SVM training on very large data sets,” Journal of Machine Learning Research, Vol. 6, pp. 363-392, Dec. 2005.
[4] I. W. Tsang, A. Kocsor, and J. T. Kwok, “Simpler core vector machines with en-closing balls,” in Proceedings of International Conference on Machine Learning, pp.911-918, Jun. 2007.
[5]N. Segata and E. Blanzieri, “Fast and scalable local kernel machines,” Journal of Machine Learning Research, Vol. 11, pp. 1883-1926, Mar. 2010.
[6]A. Oliva and A. Torralba, “Modeling the shape of the scene: A holistic represen-tation of the spatial envelope,” International Journal of Computer Vision, Vol. 42, No. 3, pp. 145-175, Jan. 2001.
[7]J. Wu, H. I. Christensen, and J.M. Rehg, “Visual place categorization: problem, dataset, and algorithm,” in Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4763-4770, Oct. 2009.
[8]D. G. Lowe, “Distinctive image features from scale-invariant keypoints,” Interna-tional Journal of Computer Vision, Vol. 60, No. 2, pp. 91-110, Nov. 2004.
[9]Y. Ke and R. Sukthankar, “PCA-SIFT: a more distinctive representation for local image descriptors,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Jul. 2004.
[10]J. M. Morel and G. Yu, “ASIFT: a new framework for fully affine invariant im-age comparison,” SIAM Journal on Imaging Sciences, Vol. 2, No. 2, Apr. 2009.
[11]H. Bay, A. Ess, T. Tuytelaars, and L. V. Gool, “SURF: speeded up robust fea-tures,” Journal of Computer Vision and Image Understanding, Vol. 110, No. 3, pp. 346-359, May 2008.
[12]K. Mikolajczyk and C. Schmid “A performance evaluation of local descriptors,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 10, No. 27, pp. 1615-1630, 2005.
[13]J. M. Morel and G. Yu, “Is SIFT Scale Invariant?” Inverse Problems and Imag-ing, Vol. 5, No. 1, Feb. 2011.
[14]D. Filliat, “A visual bag of words method for interactive qualitative localization and mapping,” in Proceedings of the IEEE/RSJ International Conference on Ro-botics and Automation, pp. 3921-3926, May 2007.
[15]P. Viola and M. J. Jones, “Robust real-time face detection,” International Journal of Computer Vision, 57(2):137-154, May. 2004.
[16]T.-J. Chin, H. Goh and J.-H. Lim, “Using densely recorded scenes for place recognition,” in Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, pp.2101-2104, Mar. 2008.
[17]A. Quattoni and A. Torralba, “Recognizing indoor scenes,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp.413-420, Jun. 2009.
[18]A. Pronobis, B. Caputo, P. Jensfelt, and H. I. Christensen, “A discriminative ap-proach to robust visual place recognition,” in Proceedings of the IEEE/RSJ Inter-national Conference on Intelligent Robots and Systems, pp.3829-3836, Oct. 2006.
[19]A. Pronobis, O. M. Mozos, and B. Caputo, “SVM-based discriminative accumu-lation scheme for place recognition,” in Proceedings of the IEEE International Conference on Robotics and Automation, pp. 522-529, May 2008.
[20]C.-J. Hsieh, K.-W. Chang, C.-J. Lin, S. S. Keerthi, and S. Sundararajan, “A dual coordinate descent method for large-scale linear SVM,” in Proceedings of Inter-national Conference on Machine Learning, pp.408-415, Jul. 2008.
[21]D. M. J. Tax and R. P. W. Duin, “Data domain description using support vectors,” in Proceedings of European Symposium on Artificial Neural Networks, pp. 251-256, Apr. 1999.
[22]D. Arthur and S. Vassilvitskii, “K-means++: The advantage of careful seeding,” in 18th Symposium on Discrete Algorithms, 2007.
[23]H. Jegou, M. Douze, and C. Schmid, “Product quantization for nearest neighbor search,” IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol. 33, No. 1, pp. 117-128, Nov. 2011.
[24]A. Pronobis and B. Caputo, “Confidence-based cue integration for visual place recognition,” in Proceedings of the IEEE/RSJ International Conference on Intel-ligent Robots and Systems, pp. 2394-2401, Oct. 2007.
[25]A. Pronobis, B. Caputo, P. Jensfelt, and H. I. Christensen, “A realistic bench-mark for visual indoor place recognition,” International Journal of Robotics and Autonomous Systems, Vol. 58, No. 1, pp. 81-96, Jan. 2010.
[26]A. Pronobis, O. M. Mozos, B. Caputo, and P. Jensfelt, “Multi-modal semantic place classification,” The International Journal of Robotics Research, Special Issue on Robotic Vision, Vol. 29, No. 2-3, pp. 298-320, Feb. 2010.
[27]R.-E. Fan, K.-W. Chang, C.-J. Hsieh, X.-R. Wang, and C.-J. Lin, “LIBLINEAR: A library for large linear classification,” Journal of Machine Learning Research, Vol. 9, pp. 1871-1874, Jun. 2008.
[28]N. Segata, “FaLKM-lib v1.0: a library for fast local kernel machines,” Technical Report DISI-09-025. 2009. Software available at http://www. disi.unitn.it/~segata/FaLKM-lib.
[29]Y. Tang, Y.-Q. Zhang, N.V. Chawla, and S. Krasser, “SVMs modeling for high-ly imbalanced classification,” IEEE Transactions on Systems, Man, and Cyber-netics, Vol. 39, No. 1, pp. 281-288, Feb. 2009.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top