臺灣博碩士論文加值系統

隨著影像行為分析、智慧型車輛、智慧型安全監控、機器視覺、人機互動等應用問題受到重視，物件追蹤這項研究議題也愈來愈熱門。但在物件追蹤的問題上，仍受到幾個變化因素限制：光線變化、姿勢變化、背景複雜、遮蔽以及相似的顏色等具挑戰性的問題。因此近年來提出的物件追蹤方法目標都在於如何克服這些因素所帶來的干擾。
在這篇論文中，我們提出了一個結合了粒子濾波器和Speeded Up Robust Features的方法。這個方法不只用了傳統的顏色資訊也利用了SURF特徵，由於SURF特徵對於光線變化、縮放以及旋轉不變性，使得我們方法的追蹤效果更加強健。由於我們針對每一個粒子都考慮了空間、色彩以及SURF特徵，估測了每一個粒子和追蹤物件之特徵的相似程度，所以我們的方法較傳統只考慮顏色資訊的方法更加強健。實驗結果也證明了我們的方法在具有挑戰性的影片中表現的不錯，除此之外，和其他篇方法相比，我們的方法在發生了部分遮蔽、和追蹤物件位於顏色相似度很高的情況下也具有競爭力。

Object tracking is important in many applications in computer vision, e.g., video analysis, intelligent vehicle, surveillance system, robot vision, human-computer interaction, and so on. This topic has received much attention in the recent decade. Although the topic of object tracking has been well studied in computer vision, it still remains challenge in varying illumination condition, noise influence, scene change, clutter background, occlusion, and similar color. Therefore, how to develop a robust method for object tracking is seriously important.
In this thesis, a novel object tracking based on particle filter and SURF feature is proposed. The proposed method uses not only color feature but also SURF feature. The SURF feature makes the tracking result more robust. Particle selection can lead to time saving. In addition, we also consider the matched particle applicable to calculating SURF weight. Owing to the color, spatial, and SURF features being adopted, this method is more robust than traditional color-based appearance model. Experimental results demonstrate the robustness and accurate tracking results with challenging sequences. Besides, the proposed method outperforms other methods during intersection of similar color and object’s partial occlusion.

摘要 II
Abstract III
Chapter 1 Introduction 1
1.1. Motivation 1
1.2. Thesis Organization 3
Chapter 2 Background and Related Works 5
2.1. Speeded Up Robust Feature (SURF) 5
2.1.1. SURF Key-points Generation 5
2.1.2. Interest Point Detection 6
2.1.3. Interest Point Descriptor 7
2.2. Tracking Using Sparse Representation 9
2.3 Object Tracking via Particle Filter Based on Foreground Extraction 10
2.3.1 Appearance model 11
2.2.1. The tracking stage 13
Chapter 3 The Proposed Method 15
3.1. Particle Selection 16
3.2. Locating Region of Foreground SURF Feature 17
3.3. Selection of Matched SURF Feature 18
3.4. Weighting of Particles 20
Chapter 4 Experimental Results 23
4.1. Experimental Parameters 23
4.2. Experimental Databases and Sequences 24
4.3. Comparison with Other Methods 25
4.3.1. PETS2001 Database 26
4.3.2. Karl-Wilhelm-Straße Database 29
4.3.3. CAVIAR Database 31
4.3.4. Girl Sequence 32
4.3.5. Faceocc 1 Sequence 34
Chapter 5 Conclusion 37
References 39

[1]S. Haner, and Yu. Gu “Combing Foreground / Background Feature Points and Anisotropic Mean Shift For Enhanced Visual Object Tracking,” International Conf. on Pattern Recognition (ICPR), 2010.
[2]Y. Liu, W. Zhou, H. Yin, and N. Yu “Tracking Based on SURF and Superpixel,” International Conf. on Image and Graphics, 2011.
[3]B. Babenko, M.H. Yang, and S. Belongie, “Robust Object Tracking with Online Multiple Instance Learning,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 33, 2011.
[4]V. Takala and M. Pietikainen, “Multi-object tracking using color, texture and motion,” in Proc. IEEE Conf. Comput. Vision Patt. Recog., Jun. 2007, pp. 1–7.
[5]J. Zhu, Y. Lao, and F.Zheng, “Object Tracking in Structured Environments for Video Surveillance Applications,” IEEE Trans. Circuits and Systems for Video Technology, vol. 20, no. 2, Feb. 2010.
[6]Y. Jhu , C. Chaung ,S. D. Lin “Multiple Object Tracking via Particle Filter Based on Foreground Extraction” , 25th IPPR Conference on computer Vision, Graphics, and Image Processing , 2012.
[7]H. Bay, A. Ess, T. Tuytelaars, and L. Van Gool, “SURF: Speeded-up robust features,” International Journal on Computer Vision and Image Understanding, vol. 110, no. 3, pp. 346-359, 2008.
[8]H. Bay, T. Tuytelaars, and L. Van Gool, “SURF: Speeded up robust features,” in Proceedings of the European Conference on Computer Vision, pp. 404-417, 2006.
[9]A. Yilmaz, O. Javed, and M. Shah. “Object tracking: A survey”. ACM Comput. Survey, 38(4), 2006.
[10] Xue Mei, Haibin Ling, “Robust Visual Tracking and Vehicle Classification via Sparse Representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 33, NO. 11, NOVEMBER 2011.
[11]K.C. Lee, and D. Kriegman. “Online Learning of Probabilistic Appearance Manifolds for Video-based Recognition and Tracking”, in Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 852-859, 2005.
[12]Q. Yu, T. B. Dinh and G. Medioni. “Online tracking and Reacquistion using co-trained generative and discriminative trackers”, in Proc. European Conf. on Computer Vision, 678-691, 2008
[13]A.Vadive1, A.K.Majumdar and S. Sural, “Performance Comparison of Distance Metrics in Content-based Image Retrieval Applications,” International Conference on Information Technology, 2003.
[14]Y. Qi, Y. Wang, “Visual Tracking With Double-Layer Particle Filter,” IEEE 11th International Conference on Signal Processing (ICSP), 2012 .
[15]T. Qin, B. Zhong, T.-J. Chin and H. Wang, “Matting-Driven Online Learning of Hough Forests for Object Tracking”, 2012 21st International Conference on Pattern Recognition (ICPR).
[16]M. Godec, P. M. Roth and H. Bischof, “Hough-based Tracking of Non-Rigid Objects,” IEEE International Conference on Computer Vision (ICCV), 2011.
[17]M. Godec, P. M. Roth and H. Bischof, “Hough-based tracking of non-rigid objects,” Computer Vision and Image Understanding, 2012.