行人偵測是近幾年來一個很新穎的研究課題。它的應用層面非常的廣泛，從一般裝設在路邊的監視器進行監視街道安全、偵測行人的人數搭配自動控制系統來節省能源，到近年來結合智慧型車輛或行車記錄器以增加行車安全，都是行人偵測的應用方向。在行人偵測中，若整個系統只使用同一種特徵系列將造成偵測率與錯誤率受到該特徵的限制，因此本論文使用兩個不同特性的分類器串聯而成，第一層使用機率模型預處理遮罩，任務就是迅速排除大量非行人區塊；第二層使用本論文提出的方向強度的混合特徵來應用於行人偵測。這個特徵混合了兩種不同的特徵，並經由AdaBoost演算法來挑選出最佳的弱分類器，以同時保有方向和強度的邊緣特性。此特徵的一大優勢就是可以利用積分圖來加速計算特徵，提升整體的處理速度，最後再依使用環境搭配上Cascade系統和Codebook背景濾除整合出一個階層式的行人偵測系統。實驗的部分，本研究使用INRIA的公開資料庫來做訓練和測試，並且比較近幾年常見用在行人偵測的特徵:Haar-like features和Edgelet features和HOG(Histogram of Oriented Gradients)搭配SVM的方法。另外在論文中也討論到解析度和行人偵測之間的關係和重要性，而進一步分析、比較在低解析度的影像中各種特徵的的優劣。從實驗數據中，證明我們提出的特徵不論在偵測速度和ROC曲線，和上述幾種特徵比較之下，皆有良好的效能，可以進一步的應用在實用的系統中。

The pedestrian detection is a popular research field for applying in the intelligent surveillance system in recent years. In this study, a hierarchical pedestrian detection system is proposed to cope with this issue. In pedestrian detection phase, if the whole system uses only one feature series, the detection rate is limited as it is solely determined by the associated feature series. To cope with this, this thesis discusses the possibility of connecting two classifiers of totally different features in series. The first classifying layer uses probability-based pedestrian mask pre-filtering, and the sum of the block filters out non-pedestrian features.The objective of the first layer is to immidiately filter out most of non-pedestrian regions while retaining as more true pedestrian as possible. The second layer exploits the proposed orientation/magnitude-based AdaBoost strong classifier. In addition, the concept of integral image is also adopted to simplify the calculations of the adopted features to increase the decision speed. Finally, according to the application environment, the system combines the cascade system and codebook model which enhances the overall processing speed to yield real-time requirement. In experimental results, some popular features such as the Haar-like feature, the edgelet feature and Histogram of Oriented Gradients (HOG) are adopted for comparison. Moreover, this study anaylized the relations and importances between resolution and pedestrian detection accuracy, and further discussing the performance in the low resolution image of the metioned features. The results demonstrate that the proposed system can yield better performance as well as high processing efficiency, and thus it can be a very competitive candidate for intelligent surveillance applications.

[1] P. Dollar, C. Wojek, B. Schiele, and P. Perona, “Pedestrian detection: a benchmark.” IEEE Conference on Computer Vision and Pattern Recognition, pp. 304-311, June 20-25, 2009.
[2] N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” in Proc. IEEE International Conference Computer Vision and Pattern Recognition, pp. 886-893, 2005.
[3] A. Mohan, C. Papageorgiou, and T. Poggio, “Example-based object detection in images by components,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 4, pp. 349-361, Apr. 2001.
[4] K. Mikolajczyk, C. Schmid, and A. Zisserman, “Human detection based on a probabilistic assembly of robust part detectors,” in Proc. European Conf. Computer Vision, pp. 69-81, 2004.
[5] K. Okuma, A. Taleghani, N. de Freitas, J. Little, and D. Lowe, “A boosted particle filter: multitarget detection and tracking,” in Proc. European Conf. Computer Vision, pp. 28-39, 2004.
[6] V. D. Shet, J. Neumann, V. Ramesh, and L. S. Davis, “Bilattice-based logical reasoning for human detection,” in Proc. IEEE International Conference Computer Vision and Pattern Recognition, 2007.
[7] I. P. Alonso, D. F. Llorca, M. A. Sotelo, L. M. Bergasa, Pedro Revenga de Toro, J. Nuevo, M. Ocana, and M. A. G. Garrido, “Combination of feature extraction methods for SVM pedestrian detection,” IEEE Trans. Intelligent Transportation Systems, vol. 8, no. 2, pp. 292-307, June 2007.
[8] A. Broggi, A. Fascioli, I. Fedriga, A. Tibaldi, and M. D. Rose, “Stereo-based preprocessing for human shape localization in unstructured environments,” in Proc. IEEE Intelligent Vehicles Symp., pp. 410-415, 2003.
[9] D. M. Gavrila, “A Bayesian exemplar-based approach to hierarchical shape matching,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 29, no. 8, pp. 1408-1421, Aug. 2007.
[10] B. Stenger, A. Thayananthan, P. H. S. Torr, and R. Cipolla, “Model-based hand tracking using a hierarchical bayesian filter,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 28, no. 9, pp. 1372-1385, Sept. 2006.
[11] T. F. Cootes, S. Marsland, C. J. Twining, K. Smith, and C. J. Taylor, “Groupwise diffeomorphic non-rigid registration for automatic　model building,” in Proc. European Conf. Computer Vision, pp. 316-327, 2004.
[12] P. Viola and M. J. Jones, “Robust real-time object detection.” International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154, 2004.
[13] Wu, B. and Nevatia, R. “Detection of multiple, partially occluded humans in a single image by Bayesian combination of edgelet part detectors,” In Proc. International Conference on Computer Vision, vol. I, pp. 90–97.
[14] P. Viola, M. Jones, and D. Snow, “Detecting pedestrians using patterns of motion and appearance,” In Proc. International Conference on Computer Vision, vol. 63, no. 2, pp. 153-161, 2005.
[15] Y. Freund and R. Schapire, “A decision theoretic generalization of on-line learning and an application to boosting.” Journal of Computer and System Sciences, vol. 55, pp. 119-139, 1997.
[16] S. Agarwal, A. Awan, and D. Roth, “Learning to detect objects in images via a sparse, part-based representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 26, no. 11, pp. 1475- 1490, Nov. 2004.
[17] B. Leibe, N. Cornelis, K. Cornelis, and L.V. Gool, “Dynamic 3D scene analysis from a moving vehicle,” in Proc. IEEE International Conference Computer Vision and Pattern Recognition, 2007.
[18] N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” in Proc. International Conference Computer Vision and Pattern Recognition, pp. 886-893, 2005.
[19] D. G. Lowe, “Distinctive image features from scale invariant keypoints.” International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004.
[20] A. K. Jain, R. P. W. Duin, and J. Mao, “Statistical pattern recognition: a review,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 1, pp. 4-37, Jan. 2000.
[21] V. N. Vapnik, The Nature of Statistical Learning Theory, Springer, 1995.
[22] C. P. Papageorgiou, M. Oren and T. Poggio, "A general framework for object detection," Sixth International Conference on Computer Vision, pp. 555-562, Jan. 1998.
[23] R. Lienhart, A. Kuranov and V. Pisarevsky, "Empirical analysis of detection cascades of boosted classifiers for rapid object detection," Pattern Recognition, vol. 2781, pp. 297-304, Sep. 2003.
[24] D. H. Kim , S. U. Jung and M. J. Chung, "Extension of cascaded simple feature based face detection to facial expression recognition," Pattern Recognition Letters, vol. 29, no. 11, pp. 1621-1631, Aug. 2008.
[25] M. Enzweiler and D. Gavrila,”Monocular Pedestrian Detection: Survey and Experiments,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31, no. 12, pp. 2179-2195, December. 2009.
[26] C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, no. 3, pp. 273-297,Sep. 1995.
[27] P. Viola and M. J. Jones, "Robust real-time face detection," International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154, May 2004.
[28] J.-M. Guo, Y.-F. Liu, C.-H. Hsia, M.-H. Shih, and C.-S. Hsu, “Hierarchical method for foreground detection using codebook model,” IEEE Transaction on Circuits and Systems for Video Technology, vol. 21, no.6, pp. 804-815, June, 2011
[29] K. Kim, T.H. Chalidabhongse, D. Harwood, and L. Davis, “Real-time foreground-background segmentation using codebook model,” Real-Time Imaging, vol. 11, no. 3, pp. 172-185, June. 2005.
[30] C. Stauffer and W.E.L Grimson, “Adaptive background mixture models for real-time tracking,” IEEE International Conference on Computer Vision and Pattern Recognition, vol.2, pp. 246–52, June, 1999.
[31] INRIA dataset: http://pascal.inrialpes.fr/data/human/
[32] SONY EVI-D70技術手冊，http://pro.sony.com/bbsc/ssr/product-EVID70P/