我們提出使用成對的區域圖形觀察與樸素貝氏分類器的圖形辨識演算法，和當今以關鍵特徵點為基準的圖形分類方法不同，如金字塔式空間性匹配演算法以及其變形方法，它是提供當今最有效的圖形分類方法之一，但他無法直覺的解釋為什麼有優良的分類效果，和此類的演算法不相同的是，我們所提出的方法是一種不會受到圖形比例、絕對位置以及旋轉所影響的演算法，我們希望藉由將成對的區域圖形觀察建立辨識模型去模擬人類的視覺系統，此模型是利用描述物件突出的觀察區域之間的關係去描繪整體物件的外觀，首先我們會提供一些背景知識，包含關鍵點特徵擷取以及不同方之間的比較，當然還有關於顯著區域的偵測方法，接下來說明關於我們設計出的模型實際的操作方法以及好處，這邊會詳細說明所有的步驟以及流程，並提供理論推導，為了驗證我們方法的正確性，我們還會論證我們的假設以及猜測，並以實驗驗證我們的數學模型，我們以目前非常多人使用的圖形資料庫 Scene-15以及Caltech-101資料庫用於演算法正確性驗證的實驗，並將正確率和詞袋模型以及金字塔式空間性匹配演算法做正確率的比較，並在最後以圖示化秀出部分實驗的結果。

We present image classification method using Naive Bayes classifier using pairwise local observations (NBPLO) based on the salient region (SR) selection and the local feature detection. Different from previous image classification algorithms, our method is a scale, translation, and rotation invariant classification algorithm. By transforming the pairwise local observations into training vectors, we may simulate the human visual system by developing the training classification model based on the neighboring relationship of the selected SRs. We verify our assumptions with Scene-15 and Caltech-101 database and compare the difference of mainstream feature point detection methods. And also compare the experiment results of bag-of-features (BoF) and SPM algorithms.

Content
Abstract I
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Previous Work 2
1.3 Overview 3
Chapter 2 Basic Assumption and System 4
2.1 Basic Assumptions and Conjectures 4
2.2 System Architecture 5
Chapter 3 Feature Extraction 7
3.1 SIFT 7
3.2 SURF 9
3.3 Dense SIFT 11
Chapter 4 Feature Quantization and Bag-of-Feature Assignment 12
4.1 K-means Algorithm 13
4.2 Online Spherical K-means Algorithm 14
4.3 Bag-of-Features Algorithm 14
4.4 Bag-of-Features Soft-Weighting Assignment 15
Chapter 5 Naive Bayes Classifier for Pairwise Local Observations 16
5.1 Salient Region Detection 16
5.2 Description Design for Detected Salient Regions 19
5.3 Local Pairwise Observation 20
5.4 Regression Model Training 21
5.5 Naïve Bayes Assumption for Object Recognition 23
Chapter 6 Implementation Details, Hypothesis Verification, and Improvements 26
6.1 The Problem of Empty Salient Regions 26
6.2 Bag of Features with Means by Kernel Weighting 27
6.3 Adjacent Local Observation with Different Scale 29
6.4 Parameters of BoF Soft-Weighting Assignment 29
6.5 The Influence of Single Observation 31
6.6 Likelihood Ratio 33
6.7 Pseudo Normalization 34
6.8 Renewed Training and Testing Process 34
Chapter 7 Experiments 37
7.1 Caltech-101 dataset 37
7.2 Scene-15 dataset 45
Chapter 8 Conclusion and Future Work 52
Reference 53

Reference
[1] Lowe, David G. "Distinctive image features from scale-invariant keypoints." International Journal of Computer Vision 60.2 (2004): 91-110.
[2] Bay, Herbert, Tinne Tuytelaars, and Luc Van Gool. "Surf: Speeded up robust features." Computer Vision–ECCV 2006. Springer Berlin Heidelberg, 2006. 404-417.
[3] Harris, Chris, and Mike Stephens. "A combined corner and edge detector." Alvey vision conference. Vol. 15. 1988.
[4] Lazebnik, Svetlana, Cordelia Schmid, and Jean Ponce. "Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories.",IEEE Conf. on Computer Vision and Pattern Recognition. Vol. 2. 2006.
[5] Yang, Jianchao, Kai Yu, Yihong Gong, and Thomas Huang. "Linear spatial pyramid matching using sparse coding for image classification." IEEE Conf. on Computer Vision and Pattern Recognition, 2009.
[6] Wang, Jinjun, Jianchao Yang, Kai Yu, Fengjun Lv, Thomas Huang, and Yihong Gong. "Locality-Constrained Linear Coding For Image Classification." IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 2010.
[7] Jiang, Zhuolin, Zhe Lin, and Larry S. Davis. "Label consistent K-SVD: learning a discriminative dictionary for recognition.", IEEE Transactions on Pattern Analysis and Machine Intelligence, 35.11 (2013): 2651-2664.
[8] Ahonen, Timo, Abdenour Hadid, and Matti Pietikainen. "Face description with local binary patterns: Application to face recognition." IEEE Trans. on Pattern Analysis and Machine Intelligence, 28.12 (2006): 2037-2041.
[9] Jabid, T. Kabir, and MH Oksam Chae. "Local directional pattern (LDP) for face recognition." Consumer Electronics (ICCE), 2010 Digest of Technical Papers International Conference on, 2010.


[10] Jia, Yangqing, Chang Huang, and Trevor Darrell. "Beyond spatial pyramids: Receptive field learning for pooled image features." IEEE Conf. on Computer Vision and Pattern Recognition (CVPR),2012.
[11] Burl, Michael C., Markus Weber, and Pietro Perona. "A probabilistic approach to object recognition using local photometry and global geometry." Computer Vision—ECCV’98. Springer Berlin Heidelberg, 1998. 628-641.
[12] Fergus, Robert, Pietro Perona, and Andrew Zisserman. "Object class recognition by unsupervised scale-invariant learning." 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 2. IEEE, 2003.
[13] Kadir, Timor, and Michael Brady. "Saliency, scale and image description." International Journal of Computer Vision 45.2 (2001): 83-105.
[14] Fei-Fei, Li, Rob Fergus, and Pietro Perona. "Learning generative visual models from few training examples: An incremental bayesian approach tested on 101 object categories." Computer Vision and Image Understanding 106.1 (2007): 59-70.
[15] Mikolajczyk, K. 2002. “Detection of local features invariant to affine transformations.” Ph.D. thesis, Institut National Polytechnique de Grenoble, France.
[16] Xiao, Jianxiong, James Hays, Krista A. Ehinger, Aude Oliva, and Antonio Torralba. "Sun Database: Large-Scale Scene Recognition from Abbey to Zoo", IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2010.
[17] Juan, Luo, and Oubong Gwun. "A comparison of sift, pca-sift and surf." International Journal of Image Processing (IJIP) 3.4 (2009): 143-152.
[18] Liu, Ce, Jenny Yuen, Antonio Torralba, Josef Sivic, and William T. Freeman. "Sift flow: Dense Correspondence across Different Scenes." Computer Vision–ECCV 2008. Springer Berlin Heidelberg, 2008. 28-42.
[19] Otero, Ives Rey, and Mauricio Delbracio. "Anatomy of the SIFT Method." (2013).
[20] Vedaldi, Andrea, and Brian Fulkerson. "VLFeat: An open and portable library of computer vision algorithms." The Int. Conference on Multimedia. ACM, 2010.
[21] Tsai, Chih-Fong. "Bag-of-words representation in image annotation: A review." ISRN Artificial Intelligence 2012 (2012).
[22] Yang, Jun, Yu-Gang Jiang, Alexander G. Hauptmann, and Chong-Wah Ngo. "Evaluating bag-of-visual-words representations in scene classification." The Int. Workshop On Multimedia Information Retrieval. ACM, 2007.
[23] Zhong, Shi. "Efficient online spherical k-means clustering. "Efficient online spherical k-means clustering.",2005 IEEE International Joint Conference on Neural Networks. (IJCNN'05), Vol. 5, 2005.
[24] Jiang, Yu-Gang, Chong-Wah Ngo, and Jun Yang. "Towards Optimal Bag-Of-Features For Object Categorization and Semantic Video Retrieval." Proceedings of the 6th ACM international conference on Image and video retrieval. ACM, 2007.
[25] Felzenszwalb, Pedro, David McAllester, and Deva Ramanan. "A discriminatively trained, multiscale, deformable part model." Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on. IEEE, 2008.
[26] Fei-Fei, Li, and Pietro Perona. "A bayesian hierarchical model for learning natural scene categories." Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on. Vol. 2. IEEE, 2005.
[27] Breiman, Leo. "Random forests." Machine learning 45.1 (2001): 5-32.
[28] Uijlings, Jasper RR, Arnold WM Smeulders, and Remko JH Scha. "What is the spatial extent of an object?." Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on. IEEE, 2009.
[29] Wang, Jingyan, Yongping Li, Ying Zhang, Chao Wang, Honglan Xie, Guoling Chen, and Xin Gao. "Bag-of-features based medical image retrieval via multiple assignment and visual words weighting." IEEE transactions on medical imaging 30, no. 11 (2011): 1996-2011.
[30] Lafferty, John, Andrew McCallum, and Fernando CN Pereira. "Conditional random fields: Probabilistic models for segmenting and labeling sequence data." (2001).
[31] Bosch, Anna, Andrew Zisserman, and Xavier Munoz. "Image classification using random forests and ferns." (2007).
[32] Oliva, Aude, and Antonio Torralba. "Modeling the shape of the scene: A holistic representation of the spatial envelope." International journal of computer vision 42.3 (2001): 145-175.
[33] Kesorn, Kraisak, and Stefan Poslad. "An enhanced bag-of-visual word vector space model to represent visual content in athletics images." Multimedia, IEEE Transactions on 14.1 (2012): 211-222.