臺灣博碩士論文加值系統

隨著門禁管制、安全監控、智慧型車輛等使用者安全性問題受到重視，人型偵測這項研究議題也愈來愈熱門。但在人型偵測的問題上，偵測效能仍受到幾個變化因素限制：光線變化、姿勢變化與背景複雜。因此近年來提出的描述算子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，我們應用結合Histogram of Oriented Gradient與韋伯定律作為特徵用於人型偵測上。這個描述算子利用韋伯定律的特性，找出強度較強的邊後，抽取出HoG特徵並使用類似HoG的統計方式抽取出韋伯定律的特徵，並將其連結作為一個偵測視窗的特徵向量。最後使用支持向量機訓練和測試。根據實驗結果，這種特徵對於INRIA資料庫擁有不錯的偵測率。又由於韋伯定律本身對噪音具有不錯的抵抗性，故這種特徵也延襲這個優點。經過實驗證明，這個特徵對於加入高斯白噪音的影像和一般現實中監視器帶有雜訊的影像，與U-HoG相比之下，仍擁有不錯的偵測率。

Human detection is essential for many applications such as surveillance and smart car. However, detecting humans in images or videos is a challenging task because of the variable appearance and background clutter. These factors affect significantly human shape. Therefore, in recent years, people are looking for more discriminative descriptors to improve the performance of human detection.
In this thesis, a robust descriptor based on HoG and Weber’s Law is proposed. Namely, the proposed descriptor is concatenated by U-HoG and histogram of Weber’s constant. Weber’s Constant has advantages such as robust to noise and detecting edge well. Because there are a large number of weak edges in the cluttered background affecting the detection result, the proposed method uses Weber’s constant to take off the weak edges. If a pixel on the weak edge, the proposed method will ignore the pixel when computing the feature. Therefore, the proposed descriptor may inherit the advantages of Weber’s Law. From the simulation results, the proposed descriptor has better performance than other comparative methods and is more robust to Gaussian white noise than U-HoG.

Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Thesis Organization 4
Chapter 2 Background 5
2.1 Weber’s Law Descriptor 5
2.1.1 Weber’s Law 5
2.1.2 Differential Excitation 6
2.1.3 Orientation 10
2.1.4 WLD Histogram 10
2.2 Histogram of Oriented Gradients Feature 11
2.3 Extended Histogram of Gradients Feature 14
2.3.1 Motivation of Ex-HoG 14
2.3.2 Formation of Ex-HoG 15
Chapter 3 The Proposed Scheme 19
3.1 Motivation of the Proposed Scheme 20
3.2 Differential Excitation 20
3.3 Weak Edge Removal Algorithm and U-HoG Feature Extraction 21
3.4 Feature Collection 23
Chapter 4 Experimental Results 25
4.1 INRIA Person Database 25
4.2 Comparison with Other Methods 27
4.3 Comparison with HoG after Adding Gaussian White Noise 31
Chapter 5 Conclusion 37
References 39

[1] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, The PASCAL Visual Object Classes Challenge 2008 Results.
[2] N. Dalal and B. Triggs, “Histograms of Oriented Gradients for Human Detection,” Proc. of CVPR 2005, vol. 1, pp.886-893, 2005
[3] S. Paisitkriangkrai, C. Shen, and Jian Zhang, “Fast Pedestrian Detection Using a Cascade of Boosted Covariance Features,” Proc. of IEEE Trans. PAMI, vol. 18, no. 8, pp. 1140-1151, 2008
[4] P. Viola and M. J. Jones, “Robust Real-Time Face Detection,” Int. Journal of Computer Vision, vol. 57, pp. 137-154, 2004
[5] O. Tuzel, F. Porikli, and P. Meer, “Human Detection via Classification on Riemannian Manifolds,” Proc. of IEEE Conf. CVPR 2007., Minneapolis, MN, pp. 1-8, 2007
[6] Z. Lin, and Larry S. Davis, “Shape-Based Human Detection and Segmentation via Hierarchical Part-Template Matching,” Proc. of IEEE Trans. PAMI, vol. 32, no. 4, pp. 604-618, 2010
[7] C. Papageorgiou, T. Evgeniou, and T. Poggio, “A Trainable Pedestrian Detection System,” Proc. of Symp. Intelligent Vehicles, pp. 241-246, 1998
[8] P. Viola and M. Jones, “Rapid Object Detection Using a Boosted Cascade of Simple Features,” Proc. of IEEE Conf. CVPR, vol. 1, pp. 511-518, 2001
[9] Y. Wu, T. Yu, and G. Hua, “A Statistical Field Model for Pedestrian Detection,” Proc. of Conf. CVPR, vol. 1, pp. 1023-1030, 2005
[10] K. Mikolajczyk, C. Schmid, and A. Zisserman, “Human Detection Based on a Probabilistic Assembly of Robust Part Detectors,” Proc. European Conf. Computer Vision, vol. 1, pp. 69-82, 2004
[11] B. Leibe, E. Seemann, and B. Schiele, “Pedestrian Detection in Crowded Scenes,” Proc. IEEE Conf. CVPR, vol. 1, pp. 878-885, 2005
[12] J. Shotton, A. Blake, and R. Cipolla, “Contour-Based Learning for Object Detection,” Proc. of IEEE Int. Conf. Computer Vision, vol. 1, pp. 503-510, 2005
[13] A. Opelt, A. Pinz, and A. Zisserman, “A Boundary-Fragment-Model for Object Detection,” Proc. European Conf. Computer Vision, vol. 2, pp. 575-588, 2006
[14] A. Mohan, C. Papageorgiou, and T. Poggio, “Example-Based Object Detection in Images by Components,” IEEE Trans. PAMI, vol. 23, no. 4, pp. 349-361, 2001
[15] A. Satpathy, X. Jiang, and H. L. Eng, “Extended Histogram of Gradients Feature for Human Detection,” Proc. of Int. Conf. on Image Processing, pp. 3473-3476, 2010
[16] C. C. Chang and C. J. Lin, LIBSVM: a library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1-27:27, 2011. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm
[17] J. Chen, S. Shan, C. He, G. Zhao, M. Pietikäinen, X. Chen, and W. Gao, “WLD: A Robust Local Image Descriptor,” IEEE Trans. PAMI, vol. 32, no. 9, pp. 1705-1720, 2010
[18] A. K. Jain, Fundamentals of Digital Signal Processing. Englewood Cliffs, NJ: Prentice-Hall, 1989
[19] D. Lowe, “Distinctive Image Features from Scale Invariant Key Points,” Int. Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004
[20] W. Jiang, “Human Feature Extraction in VS Image Using HOG Algorithm,” University of Science & Technology of China research project, 2007.