臺灣博碩士論文加值系統

本論文中發展了一個以空間和顏色為基礎的平均移動演算法。其中以空間中顏色分佈的相對資訊和顏色的特徵來定義物件的模型，並以新的相似度函數發展出新的平均移動演算法來做物件追蹤，為了要使物件追蹤的效果更穩健，針對不同的特徵做了實驗並選出使追蹤效果最好的顏色特徵，接著並在演算法中加入了以背景資訊而建立的權重值，使得演算法具有更好的穩定性。而為了解決在物件追蹤中常遇到的物件大小與方位的問題，我們使用了主成分分析的方法來估測物件的方位，並以主成分分析所延伸而來的演算法來估計物件的大小，而此方法確實可以自動更新物件的大小與方位。在最後的實驗中則可以看出此追蹤演算法可以解決部份遮蔽和物件變形的問題，且在複雜背景下仍具有良好的即時追蹤效能。

In this thesis, we propose the new mean-shift tracking algorithms based on a new similarity measure function. The joint spatial-color feature is used as our basic model elements. The target image is modeled with the kernel density estimation and we use the concept of expectation of the estimated kernel density to develop the new similarity measure functions. With these new similarity measure functions, two new similarity-based mean-shift tracking algorithms were derived. To enhance the robustness, we add the weighted-background information to the proposed mean-shift tracking algorithm. In order to solve the deformation problem, the principal component analysis method is used to update the orientation of the tracking object, and a simple method is elaborated to monitor the scale of the object. The results of the experiments show that the new similarity-based tracking algorithms are real-time and can track the moving object correctly, and update the orientation and scale of the object automatically.

摘要 i
ABSTRACT ii
誌謝 iii
Contents iv
List of Tables vi
List of Figures vii
Chapter 1. Introduction 1
1.1 MOTIVATION AND OBJECTIVE 1
1.2 LITERATURE REVIEW 2
1.3 THESIS SUBJECT AND CONTRIBUTION 4
1.4 OUTLINES OF THESIS 5
Chapter 2. Traditional Mean-Shift Tracking Algorithm 6
2.1 INTRODUCTION 6
2.2 TARGET REPRESENTATION 6
2.2.1 Model Representation 7
2.2.2 Candidate Representation 8
2.3 SIMILARITY BASED ON BHATTACHARYYA COEFFICIENT 9
2.4 TRADITIONAL MEAN-SHIFT TRACKER 10
2.5 MEAN-SHIFT TRACKING ALGORITHM PROCEDURE 13
Chapter 3. Spatial-Color Mean-Shift Object Tracking Algorithm 15
3.1 INTRODUCTION 15
3.2 MODEL DEFINITION 15
3.2.1 Paper Survey about Spatiogram 16
3.2.2 A Joint Spatial-Color Feature Model 17
3.3 PAPER SURVEY ABOUT NEW SIMILARITY MEASURE 18
3.4 SPATIAL-COLOR MEAN-SHIFT OBJECT TRACKING ALGORITHM 19
3.4.1 Kernel Density Estimation of the Model Image 19
3.4.2 Similarity Measure Function 20
3.4.3 Spatial-Color Mean-Shift Tracker 21
3.4.4 Another Derivation of the New Mean-Shift Tracker 22
3.4.5 Spatial-Color Mean-Shift Tracking Procedure 25
3.5 CHOICE OF THE COLOR FEATURE SPACE 27
3.6 BACKGROUND-WEIGHTED INFORMATION 28
3.7 UPDATE OF SCALE AND ORIENTATION 29
3.7.1 Introduction of Principal Component Analysis 30
3.7.2 Orientation Selection by Principal Component Analysis 31
3.7.3 Adding Weighted-Background Information 33
3.7.4 Scale Selection 34
3.8 SUMMARY 35
Chapter 4. Experiment Results 38
4.1 EXPERIMENT ILLUSTRATION 38
4.2 SPATIAL-COLOR MEAN-SHIFT TRACKERS WITH RGB FEATURE 39
4.2.1 Face Sequence 39
4.2.2 Cup Sequence 41
4.2.3 Walking Girl Sequence 43
4.3 SPATIAL-COLOR MEAN-SHIFT TRACKERS WITH NORMALIZED FEATURE 44
4.3.1 Face Sequence 44
4.3.2 Cup Sequence 48
4.3.3 Walking Girl Sequences 52
4.4 SPATIAL-COLOR MEAN-SHIFT TRACKERS WITH NORMALIZED FEATURE AND WEIGHTED INFORMATION 54
4.4.1 Face Sequence 55
4.4.2 Cup Sequence 61
4.4.3 Walking Girl Sequence 66
4.5 SPATIAL-COLOR MEAN-SHIFT TRACKERS WITH SCALE AND ORIENTATION 70
4.5.1 Walking Person Sequence 70
4.5.2 Surveillance Sequences 74
4.6 PERFORMANCE ANALYSIS 78
Chapter 5. Conclusion and Future Work 82
References 83

[1] D. Comaniciu, V. Ramesh, and P. Meer, “Kernel-based object tracking,” IEEE Trans. On Pattern Analysis and Machine Intelligence, 25(5):564-577, May 2003.
[2] F. Porikli, O. Tuzel, “Multi-kernel object tracking,” IEEE International Conference on Multimedia and Expo, pp. 1234-1237, 2005.
[3] V. Parameswaran, V. Ramesh, and I. Zoghlami, “Tunable kernels for tracking,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol.2, pp.2179-2186, 2006.
[4] R. Collins, “Mean-shift blob tracking through scale space”, Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2003.
[5] K. She, G. Bebis, H. Gu, and R. Miller, “Vehicle tracking using on-line fusion of color and shape features”, Proc. IEEE Conf. on Intelligent Transportation Systems, 2004.
[6] Q. Zhao and H. Tao, “Object tracking using color correlogram,” Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2005.
[7] S. Birchfield and S. Rangarajan, “Spatiograms versus histograms for region-based tracking,” Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2005.
[8] H. Zhang, Z. Huang, W. Huang, and L. Li, “Kernel-based method for tracking objects with rotation and translation,” Proc. International Conf. on Pattern Recognition, 2004.
[9] Z. Zivkovic and B. Krose, “An em-like algorithm for color-histogram-based object tracking,” Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2004.
[10] F. Porikli and O. Tuzel, “Object tracking in low-frame-rate video,” Proc. PIE/EI—Image and Video Communication and Processing, San Jose, CA, 2005.
[11] T. Lindeberg, “Feature detection with automatic scale selection,” International Journal of Computer Vision, vol. 30(2), pp.79-116, November 1998.
[12] C. Yang, R. Duraiswami, and L. Davis, “Efficient mean-shift tracking via a new similarity measure,” IEEE Conf. on Computer Vision and Pattern Recognition, vol. 1, pp. 176-183, 2005.
[13] T. Hastie, R. Tibshirani, andJ. Friedman, “The elements of statistical learning,” Springer, 2001.
[14] http://www.cs.technion.ac.il/~amita/fragtrack/fragtrack.htm
[15] http://groups.inf.ed.ac.uk/vision/CAVIAR/CAVIARDATA1/
[16] D.W. Scott, “Multivariate Density Estimation,” New York: Wiley, pp.24-26, 1992.
[17] D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” Best Paper Award, IEEE Conf. Computer Vision and Pattern Recognition, Hilton Head Island, South Carolina, vol.2, pp.142-149, 2000.
[18] D. Comaniciu, P. Meer, “Mean shift: a robust approach toward feature space analysis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.24 n.5, pp.603-619, May 2002