跳到主要內容

臺灣博碩士論文加值系統

(44.222.218.145) 您好!臺灣時間:2024/02/26 23:46
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:張緯浩
研究生(外文):Wei-Hao Chang
論文名稱:基於外觀的高效率物件辨識方法
論文名稱(外文):An Efficient Method for Appearance-based Object Recognition
指導教授:洪一平洪一平引用關係
指導教授(外文):Yi-Ping Hung
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:資訊工程學研究所
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:2007
畢業學年度:95
語文別:英文
論文頁數:44
中文關鍵詞:外觀物件辨識模式剃除特徵抽取模式匹配
外文關鍵詞:appearanceobject recognitionpattern rejectionfeature extractionpattern matchingWalsh-Hadamard
相關次數:
  • 被引用被引用:0
  • 點閱點閱:441
  • 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:0
物件辨識一直是過去電腦視覺領域想要解決的重要目標。為了讓物件辨識技術 可以更快速地應用在我們的生活當中,我們發展出了一套高效率的物件辨識方法 可以讓電腦系統經由快速的訓練過程,使電腦系統可以快速地進行物件辨識工作 並達到很好的辨識率。

我們利用 Image Retrieval 的概念來設計所提出的物件辨識技術,將欲辨識的 物件影像轉換為特徵向量,進而在資料庫中找尋與其最為相似的特徵向量集合, 統計出現次數最多的物件類別即為辨識結果。

為了達成我們提出的辨識方法,我們提出了以下幾種新的技術:一個階層化剔 除樹 (Hierarchical Rejection Tree),這個技術可以快速地尋找與被辨識物件 影像最為相似的特徵向量,在這個技術當中,我們改良Hel-Or 2005 年9 月在PAMI 發佈的研究成果(Projection Scheme) [3]。我們所提出的Hierarchical Rejection Tree 提供了高效率的搜尋機制,除了用在本論文的應用之外,也可以 應用在需要高維度特徵空間的搜尋應用加速上。此外 我們也發展了兩個有效率的 特徵萃取方法(feature extraction method):Hue mean/Intensity DCT-WH 影像 特徵及Hue mean/Hue histogram-WH 影像特徵,利用影像的頻譜特性及色彩分布 當作影像特徵,將其投影至Walsh-Hadamard basis vectors 並加入色彩資訊(色 調均值)以增進影像特徵的代表性。

結合上述的新技術,我們發展出一套兼具速度與準確度的物件辨識方法 並經由 實驗數據証明結合這些技術所發展出來的物件辨識系統具有快速訓練,快速辨識 及高辨識率等特色。
In computer vision, the ways to make computers being capable of seeing and understanding the world have been intensely studied more than three decades. In this thesis, an efficient method for appearance-based object recognition is proposed. The proposed method including both the training phase and recognition phase is very efficient.

The concept of image retrieval is applied in our method. It tries to find the most similar feature vector set in image databases and classify the testing image by using the most frequent class label in the feature vector set.

For accomplishing the proposed object recognition method, we developed several techniques: an efficient pattern rejection scheme - Hierarchical Rejection Tree is first proposed here. Hierarchical rejection tree can find the most similar feature vector in database efficiently. Based on Hel-Or’s Projection Scheme [3], we enhanced its performance by iterative indexing tree structure. The experimental result shows the performance of Hierarchical Rejection Tree is faster 2.6 times than projection scheme. In this thesis, two new effective feature extraction methods are also be introduced. For an image, the hue mean/intensity DCT-WH image feature captures color information and DCT spectrum as its characteristics. For an image, the other feature extraction, the hue mean/hue histogram-WH image feature captures color information and color distribution as its characteristics.

We combine the Hierarchical Rejection Tree technique with feature extraction methods to develop our appearance-based object recognition system. Three image databases which contain large number of object images are used as testing databases. The experimental results show the proposed method has the following features: fast training speed, fast recognition speed and high recognition rate.
Contents
Chapter 1 Introduction 1
Chapter 2 Related Works 3
Chapter 3 Overview of Proposed Method 5
Chapter 4 Feature Extraction Methods 9
4.1 Hue mean/Intensity DCT-WH Image Feature 10
4.2 Hue mean/Hue histogram-WH Image Feature 14
Chapter 5 Hierarchical Rejection Tree 17
5.1 The Projection Scheme 18
5.2 Similarity Measurement 19
5.3 The Hierarchical Rejection Tree 21
5.3.1 Overview 21
5.3.2 Hierarchical Rejection Tree - Building Process 26
5.3.3 Hierarchical Rejection Tree – Querying Process 26
Chapter 6 Experimental Results 29
6.1 Testing Image Databases 30
6.1.1 Amsterdam Library of Object Images (ALOI) 31
6.1.2 Columbia Object Image Library (COIL-100) 32
6.1.3 Museum Artifacts Object Movie database 33
6.2 Experimental Results 34
6.2.1 Experimental result of ALOI database 34
6.2.2 Experimental result of COIL-100 database 34
6.2.3 Experimental result of Museum Artifacts Object Movie database 35
6.2.4 Fast Rejection 36
6.2.5 Hierarchical Rejection Tree and Projection Scheme 37
6.2.6 Performance of proposed method 38
Chapter 7 Conclusion and Future Works 39
7.1 Conclusion 39
7.2 Future Works 40
References 41

List of Figures
Figure 1. Overview of training process 6
Figure 2. Overview of recognition process 7
Figure 3. The feature extraction procedure of Hue mean/Intensity DCT-WH Image Feature 10
Figure 4. 2D Walsh-Hadamard transformation 12
Figure 5. The feature extraction procedure of Hue mean/Hue histogram-WH image feature 14
Figure 6. Projection of difference vector on projection vectors [3] 18
Figure 7. Applying Projection Scheme for similarity measurement 20
Figure 8. Compare linear search with Tree search 22
Figure 9. Projection Scheme, Hierarchical Rejection Tree and Result list 24
Figure 10. The Jellyfish Query Model 28
Figure 11. The user interface of our implementation 30
Figure 12. Examples of Amsterdam Library of Object Images (ALOI) 31
Figure 13. Examples of Columbia Object Image Library (COIL-100) 32
Figure 14. Examples of Museum Artifacts Object Movie database 33
Figure 15. Experimental result of fast rejection 36
Figure 16. Rejection effect from basis 1 to basis 11 37

List of Tables
Table 1. Experimental result of ALOI database 34
Table 2. Experimental result of COIL-100 database 35
Table 3. Experimental result of Museum Artifacts Object Movie database 36
Table 4. Performance comparison between Hierarchical Rejection Tree and Projection Scheme 38
Table 5. Performance summarization of proposed method 38
[1] J. M. Geusebroek, G. J. Burghouts, and A. W. M. Smeulders, “The Amsterdam library of object images,” Int. J. Computer Vision, vol. 16, no. 1, pp. 103-112, Sep. 2004.
[2] Y. Hel-Or, and H. Hel-Or, “Real time pattern matching using projection kernels,” Proc. of Ninth IEEE International Conference on Computer Vision, vol. 1, Oct. 2003, pp. 1486 – 1493.
[3] Y. Hel-Or, and H. Hel-Or, “Real time pattern matching using projection kernels,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 27, No. 9, Sep. 2005, pp. 1430 - 1445.
[4] Z. Pan, A. G. Rust, H. Bolouri, “Image Redundancy Reduction for Neural Network Classification using Discrete Cosine Transforms,” Proceedings of the IJCNN, vol. 3, pp. 149-154, 2000.
[5] S. K. Nayar, S. A. Nene, and H. Murase, “Real-time 100 object recognition system,” in Proc. Of ARPA Image Understanding Workshop, February 1996.
[6]S. A. Nene, S. K. Nayar, and H. Murase, “Columbia Object Image Library (COIL-100),” Techn. Rep. No. CUCS-006-96, dept. Comp. Science Columbia University, 1996.
[7] A.J. Ahumada, “Computational Image Quality Metrics: A Review,” Proc. Soc. Information Display Int’l Symp., vol. 24, pp. 305- 308, 1998.
[8] A. M. Eskicioglu, and P. S. Fisher, “Image Quality Measures and Their Performance,” IEEE Transactions on Communication, vol. 43, no. 12, pp. 2959–2965, December 1995.
[9] B. Girod, “What’s wrong with Mean-Squared Error?” Digital Images and Human Vision, chapter 15, pp. 207-220. MIT Press, 1993.
[10] S. Santini and R. Jain, “Similarity Measures,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 21, no. 9, pp. 871-883, Sept. 1999.
[11] M. A. Turk, and A. P. Pentland, “Face Recognition Using Eigenfaces”, Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 586-591, June 1991.
[12] S. Baker and S. K. Nayar, “Pattern Rejection,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 544-549, 1996.
[13] M. Elad, Y. Hel-Or, and R. Keshet, “Pattern Detection Using Maximal Rejection Classifier,” Proc. Int’l Workshop Visual Form, pp. 28-30, May 2000.
[14] http://en.wikipedia.org/wiki/Main_Page
[15] R. Gonzalez, and R. Woods, Digital Image Processing 2/E. Reading, MA: Addison-Wesley, 2001.
[16] T. Gevers, and A. W. M. Smeulders, “Color based object recognition,” Pattern Recognition 32, pp. 453–464, 1999.
[17] M. Stricker and M. Orengo, “Similarity of color images,” In Proc. SPIE Storage and Retrieval for Image and Video Databases, 1995.
[18] H. Kitajima, “Energy Packing Efficiency of the Hadamard Transform,” IEEE Trans. Comm., pp. 1256-1258, 1976.
[19] M. J. Swain, D. H. Ballard, “Color indexing,” Int. J. Computer Vision, vol. 7, no.1, pp. 11-32, 1991.
[20] Y. P. Hung, C. S. Chen, Y. P. Tsai, and S. W. Lin, “Augmenting Panoramas with Object Movies by Generating Novel Views with Disparity-Based View Morphing,” Journal of Visualization and Computer Animation, vol. 13, pp. 237-247, 2002.
[21] http://en.wikipedia.org/wiki/Hierarchical_Temporal_Memory
[22] T. Poggio, and S. Edelman, “A network that learns to recognize three-dimensional objects,” Nature 343, pp. 263-266, 1990.
[23] D. Roth, M. H. Yang, and N. Ahuja. “Learning to recognize objects,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 724–731, 2000.
[24] M. Pontil and A. Verry. “Support vector machines for 3D object recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 20, pp. 637–646, 1998.
[25] J. Hawkins, and S. Blakeslee. On Intelligence. Henry Holtand Company, LLC, New York, 2004.
[26] D. George, and J. Hawkins. “A hierarchical Bayesian model of invariant pattern recognition in the visual cortex,” In Proceedings of the International Joint Conference on Neural Networks. 2005.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top